Contrast agents for cardiovascular optical imaging at

UNIVERSIDAD AUTÓNOMA DE MADRID FACULTAD DE CIENCIAS

DEPARTAMENTO DE FÍSICA DE MATERIALES

Contrast agents for cardiovascular optical imaging at

molecular level

Doctor thesis

Jie Hu

Madrid, 2018

UNIVERSIDAD AUTÓNOMA DE MADRID FACULTAD DE CIENCIAS

DEPARTAMENTO DE FÍSICA DE MATERIALES

Contrast agents for cardiovascular optical imaging at

molecular level

Doctor thesis

Jie Hu

Supervisor:

Prof. JOSÉ GARCÍA SOLÉ

Dr. EMMA MARTÍN RODRÍGUEZ

Tutor:

Prof. DANIEL JAQUE GARCÍA

Doctor program:

Física de la Luz y la Materia

Madrid, June, 2018

To my parents

V

Acknowledgements

Like Camino de Santiago, you finished your camino when you arrived at Catedral de Santiago

de Compostela, but the end is not there, the end is in the west of the mainland, where you can see

the sunset on the sea…

During these three years, I have learned a lot about optics, such as light source, objectors, mirrors,

lenses, filters, CCD and InGaAs detector. In addition, some knowledge about cells, arteries and

animals. Thanks for the FIG members helping me to learn these things. Even more, I have seen

the cardiovascular surgery in human bodies. I cann’t forget the moment, when Pepe, Dani and me

were in hospital to do my first work on cardiovascular disease. At that moment, we didn’t know

how to do the experiment, we just kept discussing with the doctors and playing with gold

nanoparticle and OCT catheter. That was the first time I stayed in the surgery room, and played

with clinical techniques. I want to say the first step of my Ph.D was started from there, and that

experience keep me going on and on in cardiovascular disease.

Foremost, I would like to thank my supervisor Dani, who made my dream to study abroad come

ture, and makes this thesis a possibility. I am very appreciate him for teaching me how to do

experiments, analyze data and draw sexy figures. He is the man who helps me to get new samples

to do research, always pushes me to work and stays in the lab work with me. I still remember his

effort in helping me to get collaboration with other groups, providing me such amount of

opportunities to attend conferences, and encouraging me to talk with other researchers. Sometimes,

we went to the right place but wrong conference, he said I deserved to enjoy the scenery and relax.

In case of football, he is a fans of Atlético Madrid, but he also supports my team Real Madrid. I

never forgot the time we watched the final Champions League between Real Madrid and Liverpool

on the plane in Poland.

I also thank my co-supervisor Pepe, for untiring discussions about theory and data analyses, and

effort in improving my oral presentation. I want thanks him for carefully revising my thesis, and I

have learned a lot in writing. I remember for the first work, we have good results, but we didn’t

know how to analyze them, and what is the main focus of that work. He kept giving me the ideas

VI

to deal with those data. In the discussion about gold nanoshells for cells imaging, we didn’t know

what were the red spots in the HeLa cells, and how the incubation time affected the number of

GNSs in HeLa cells. We have done a lot of work, but the results were still not very good, and we

didn’t know what was the next step, at that moment, we lost temper. But finally we got the paper

published in Nano Research. Then I learnt how to take the responsibility to do the experiment, and

from then on, strangely, I prefer to have an argument with him, because I know, only in this way,

I can learn things. As we know, he is a Madridista and he likes football very much. I cann’t forget

the precious moment I played football with him in the field, and also the moment I watched the

Champions league between Real Madrid and PSG at his house. He was so kind, and often, he lent

me his season card to watch La Liga in Santiago Bernabéu stadium. You can not imagine the

jealous from my friends.

My sincerest thanks extend to my co-supervisor Emma, she is such a nice woman, without her

help, the thesis would be impossible. She spent almost all the time to take care of my thesis. It was

her who always came my office, and gave me intellectual suggestions on writing the thesis. At this

moment, I would like to thanks her husband, Dirk. Both of them teach me how to do the synthesis,

this would be helpful for my future. Without helping from them, the functionalization of

nanoparticles for bioapplication would be impossible. Thanks them for colorful topics in every

day’s coffee time, we shared photos, we discussed what our plan for the weekend and what are

places of interests to visit. Sometimes, we complained about the long under review time for a paper.

Thanks them and their family for the invaluable gift for my birthday from Netherland, they always

made me feel working at home. I really like their daughter, Almudela, who is almost 1 year old

and with big blue eyes. Every time, she appears in our office, it seems the angle is there, you cann´t

pay no attention to her, she always make your life much more relax. I remember there was a day,

when I said goodbye to her, she feels very sad, I don´t know if it will happen again when I leave

here.

I would also like to thank Patri, actually I would like to say she is my professional optical

technique. She helped me to build dark field microscopy setup. Every time when I didn´t know

how to solve the problem in Andor, she always known how to manage it. I cann´t forget the bad

thing I did, the first day when I was in the lab, I left the lab without switching off the Ti: Sapphire,

and she helped me to switch it off. From then on, I started try to build a good habit in doing

VII

experiment. She is very good at in sport activities, and she pushed me to practice Karate. Now I

am much more stronger than before.

I thank Lucía for listening my complains and encouraging me all the time. Without sharing the

experience with you, I will be completely lost in doing research. In particularly, she is my translator

in the lab, it seems only she can understand my Chinese-English. Indeed, I also thanks for her kind

remind in organizing everything, and make things much easier in the lab. I want to mention the

medieval festival we went in Alcalá de Henares, that was my first time to visit a place with foreign

friends, thanks for her interesting instruction about the history and delicious food there.

I want to thank Paloma, she is a girl who works very hard and almost the last one left the lab,

and she never complains. I always take her as an example and encourage myself to work like her.

I thank Paco for offering me the cells to do the experiment, without these cells, in vitro

experiment would be impossible. I remember the first time we used OCT catheter to detect HeLa

cells, which was an unexpected finding, and makes my thesis much more interesting now.

I also thank to Nuria for offering me the arteries to do the ex vivo experiment. She was always

available whenever Dani ask her to provide the arteries.

I would like to express my gratitude to all current and former members of FIG group, in

particular, Harrison, Ruiyun. Blanca, Antonio, Erving and Uéslen, thanks for their kind help. They

made my life much more easier than expected.

I thank to David and Alejandro, who have shared the office with me. They are so kind heart

guys and always give me a hand when I was suffering. They always smile, bring the light to me,

and make me happy. Fortunately, we have shared a good moment in the football field.

I am also very grateful to our collaborators. Thanks Dr. Fernando Rivero, Dr. Río Aguilar Torres

and Prof. Fernando Alfonso (Hospital Universitario de la Princesa) for providing us the OCT

equipment, and pushing us to keep on through time to time of meetings. Thanks Prof. Pilar Rivera

Gil, Paula and Dr. Dionysia Tsoutsi (Universitat Pompeu Fabra) for providing us gold nanostars,

and teaching me how to do flowing experiment. Thanks José A. Sánchez-Gil and Diego Romero

(Instituto de Estructura de la Materia (IEM-CSIC)) for providing the simulation data of extinction,

VIII

scattering and absorption spectra of gold nanoparticles. Thanks Prof. Horacio Lamela Rivera, Dr

Luca llegio, and Dr Daniel Gallego (Universidad Carlos III de Madrid) for their help in doing

photoacoustic experiment.

I would like to thank my football team in Getafe and UAM, and my basketball team in UAM.

Thanks them for making my spare time much more interesting. These famous players in football

teams are Tian Bing, Ye Tang Ye, Zhang Lu, Deng Xiangxing, Li Wenda, and Wu Beilun. In

basketball teams, they are Yansheng, Liu Jian, Huayu, Qinglong, Tiancun, Xiyan, Shaoliang,

Zhaoyu, Lizhi, Li Chuan, Yuwen, Gaole, Ingyu, Yuhao and Xuejun. I would like to thanks my

classmates who were also supported by CSC, we have a very good memory during my Ph.D studies.

They are Li Xiong, Jinbao, Yuyang, Chen Yu, Jingya, Fu Can. Particularly, I would like to thanks

Deng Xiangxing, who should be mentioned above. He is my good friend, we have many nice trips

and we always encourage each other to work hard. I also want to thank Junqing, Longfei and

Yukun, we have enjoyed a lot of Chinese restaurant in Madrid and shared the experience in finding

a job. I want to extend my grateful to my friends in China, who always take care of me, and enjoy

the sceneries I shared with them, and shared their colorful life with me, they are Wang Han, Huang

Jianhui, Gong Guoliang, Lu Dasheng, Chen Sijie, Han Xuequn, Wang Zhenqiang, Zhang Jifa, Du

Binbin…

Thanks to all the foundations for supporting my life and study during these three years. They

are the Ph.D grant from China Scholarship Council (CSC, No. 201506650003) and Spanish

Ministerio de Economía y Competitividad under Project (No. MAT2016-75362-C3-1-R).

Finally, I want to thanks my family. It has been for 10 years since I sat in the classroom, and

attent the college entrance examination. At that time, I promised to myself that I need to finish the

exam very well in order to attend a good university and then I can support my family. But 10 years

later, I am still a student, and I cann’t do anything for my family. They always support me without

any conditions. When they were suffering, they tried to avoid to tell me in order to make my focus

on the study. Especially this year, all of them are suffereing injury in their left hands, and cann’t

work for a long time. Due to lack of money, my father has to go back to work without total recovery,

and my mother has to stay at home and make food for herself. I can not imagine what is the

situation now with my parents, and I can do nothing for them. They never complain it to me and

IX

just keep asking me to focus on experiment and thesis. Thanks for my sister and sister in law to

take care of them, and supporting my study abroad. Thanks for my nephew, he is the guy who

always make me happy. I would like to thanks my grandma, when I was a child, it was her who

got up very early and prepared the breakfast for me. Even she was dead, she is still aliving in my

heart.

To all of you, thanks.

Contents Acknowledgements .................................................................................................................... V

Summary .................................................................................................................................. XV

Resumen ................................................................................................................................ XVII

Chapter 1 Introduction and motivation ....................................................................................... 1

1.1 Cardiovascular diseases .................................................................................................... 2

1.2 Cardiovascular imaging techniques based on light ........................................................... 5

1.2.1 Optical coherence tomography .................................................................................. 5

1.2.2 Photoacoustic imaging ............................................................................................... 8

1.2.3 Fluorescence imaging............................................................................................... 10

1.3 Contrast agents ................................................................................................................ 12

1.3.1 Gold nanoparticles (GNPs) ...................................................................................... 13

1.3.2 Quantum dots (QDs) ................................................................................................ 20

1.4 Targeting strategy............................................................................................................ 24

1.4.1 Passive targeting strategy ......................................................................................... 24

1.4.2 Active targeting strategy .......................................................................................... 25

1.5 Motivation ....................................................................................................................... 27

Chapter 2. Experimental techniques ......................................................................................... 29

2.1. Preparation of nanoparticles ........................................................................................... 30

2.1.1. Synthesis and functionalization of GNPs ................................................................ 30

2.1.2. Synthesis and functionalization of QDs .................................................................. 30

2.2. Basic characterization .................................................................................................... 31

2.2.1. Transmission electron microscopy (TEM) ............................................................. 31

2.2.2. Total reflection X-ray fluorescence (TXRF) ........................................................... 31

2.2.3. Extinction spectroscopy .......................................................................................... 31

2.2.4. Photoluminescence spectroscopy ............................................................................ 32

2.2.5. Dark field microscopy (DFM) imaging system ...................................................... 32

2.2.6. Photothermal method and experiments ................................................................... 34

2.2.7. Optical coherence tomography ............................................................................... 37

2.2.8. Infrared fluorescence imaging ................................................................................ 44

2.2.9. Photoacoustic experiment ....................................................................................... 45

2.3. In vitro experiment ......................................................................................................... 46

2.3.1. HeLa and Jurkat cells; culture and incubation with Gold Nanoshells (GNSs) ....... 46

2.3.2. Human Mammary Epithelial Cells culture, activation and evaluation of molecular

expression ............................................................................................................................. 46

2.3.3. Cell viability study .................................................................................................. 48

Chapter 3. Gold nanoparticles in water dispersion visualization by OCT ................................ 51

3.1. Introduction .................................................................................................................... 52

3.2. Characterization of GNPs .............................................................................................. 53

3.3. Static OCT imaging experiments ................................................................................... 56

3.4. Three dimensional imaging of GNPs ............................................................................. 62

3.5. OCT imaging of flowing GNPs ..................................................................................... 64

3.6. Conclusions .................................................................................................................... 65

Chapter 4. Experimental evaluated scattering properties of gold nanoparticles ....................... 67

4.1. Introduction .................................................................................................................... 68

4.2. Characterization of the GNPs......................................................................................... 69

4.3. Experimental determination of infrared scattering efficiency ....................................... 70

4.4. Numerical calculations and comparison with experimental data ................................... 74

4.5. Infrared scattering experiments ...................................................................................... 78

4.5.1. Infrared dark field microscopy ................................................................................ 79

4.5.2. Intravascular optical coherence tomography .......................................................... 80

4.5.3. Optoacoustic experiments ....................................................................................... 82

4.6. Conclusions .................................................................................................................... 84

Chapter 5. GNSs as contrast agent for cell imaging and tissue contrast enhancement by OCT87

5.1. Introduction .................................................................................................................... 88

1.1. Static OCT imaging of HeLa and Jurkat cells incubated with GNSs ............................ 89

1.2. Additional confirmation of the internalization of the GNSs into the cells .................... 92

1.3. Incubation efficiency of GNSs incubated into the cells ................................................. 94

1.4. Contrast enhancement at cellular and tissue level.......................................................... 99

1.5. Internalization of GNSs in HMEC-1 cells under flow conditions ............................... 102

1.6. Conclusions .................................................................................................................. 103

Chapter 6. PbS QDs as contrast agents for cardiovascular bimodal imaging ......................... 105

6.1. Introduction .................................................................................................................. 106

6.2. Characterization of PbS QDs ....................................................................................... 107

6.3. Bimodal OCT and fluorescence imaging of colloidal QDs ......................................... 109

6.4. OCT and fluorescence bimodal imaging in tissues ...................................................... 114

6.5. OCT and fluorescence bimodal imaging in rabbit artery ............................................. 117

6.6. Conclusions .................................................................................................................. 119

Chapter 7. Conclusions and perspectives ................................................................................ 121

7.1. GNSs, best OCT contrast agent ................................................................................... 122

7.2. Single GNSs visualization by IV-OCT ........................................................................ 122

7.3. GNSs can act as contrast agents for cell imaging and tissue contrast enhancement by IV-

OCT......................................................................................................................................... 122

7.4. PbS QDs could act as a dual contrast for cardiovascular imaging ............................... 122

7.5. Future perspectives....................................................................................................... 123

Chapter 8. Conclusiones y Perspectivas futuras (ES) ............................................................. 125

8.1. GNSs, mejor agente de contraste para OCT ................................................................ 126

8.2. Visualización de GNSs individuales por IV-OCT ....................................................... 126

8.3. Las GNSs pueden actuar como agentes de contraste para imagen celular y aumento del

contraste en tejidos por IV-OCT ............................................................................................. 126

8.4. QDs de PbS pueden actuar como agentes de contraste duales para imagen cardiovascular

................................................................................................................................................. 126

8.5. Perspectivas futuras ...................................................................................................... 127

References ............................................................................................................................... 129

XV

Summary

In this thesis, contrast agents for cardiovascular optical imaging at molecular level have been

investigated. The investigation is mainly based on intravascular optical coherence tomography

(IV-OCT), and the combination of IV-OCT and fluorescence imaging. Cardiovascular disease is

the main cause of the death in the world. Until now, IV-OCT as an minimally invasive technique

has been used for in vivo evaluation of unstable plaque in atherosclerosis and stent placement. The

development of nanometric contrast agents for IV-OCT would help to visualize the atherosclerosis

process at molecular level. By combining the IV-OCT with fluorescence imaging, the contrast

agent would be fully used and provide a high resolution of cardiovascular imaging.

First, the contrast enhancement provided by different kind of gold nanoparticles have been

investigated by IV-OCT. The function of IV-OCT is determined by the static and pull back option

in IV-OCT system. IV-OCT allows the detection of single gold nanoparticles due to its high

sensitivity.

Second, due to the IV-OCT signal is based on the scattering ability of the sample, the scattering

properties of gold nanoshells (GNSs), nanorods, and nanostars have been systematically

investigated by numerical simulation, photo-thermal method, infrared dark field microscopy,

photoacoustic and IV-OCT. GNSs have been determined to be a good scatterers for IV-OCT.

Third, the GNSs as contrast for cell and tissue imaging by IV-OCT have been investigated. The

incubation of GNSs in HeLa and Jurkat cells produced a high OCT contrast enhancement. The

internalized of GNSs in these cells have been confirmed by dark field microscopy and TEM. A

OCT contrast enhancement after injected of GNSs in tissue is also obtained.

Finally, the quantum dots (QDs) which can be excited by IV-OCT laser and show an emission

in third biological window have been investigated. Due to the high sensitivity of IV-OCT, these

QDs also give OCT contrast enhancement. The QDs used for OCT and fluorescence dual imaging

XVI

has been systematically investigated in colloidal, tissue and in the artery. From the fluorescence

imaging of QDs excited by the pull-back IV-OCT laser could provide the morphology of artery.

In conclude, the investigated GNSs and QDs would be very useful for cardiovascular optical

imaging at molecular level. The functionalization of these nanoparticles would make them

specially target to the cardiovascular disease site, and provide a high resolution imaging for

visualization of atherosclerosis process.

Key words: cardiovascular disease, atherosclerosis plaque, IV-OCT, fluorescence imaging, gold

nanoshells, quantum dots.

XVII

Resumen

En esta tesis se han investigado diversos agentes de contraste para imagen cardiovascular a nivel

molecular. Las investigaciones se basan fundamentalmente en la técnica de Tomografía Óptica

Coherente Intravascular (TOC-IV). Actualmente, las enfermedades cardiovasculares son la causa

de la mayoría de los decesos en el mundo. Desde hace unos años, la TOC-IV se ha venido

utilizando como técnica mínimamente invasiva para la implantación de stents y la detección y

evaluación de placas de ateroesclerosis. El desarrollo de agentes de contraste nanométricos,

ayudaría a detectar dichas placas a nivel molecular. Además, si se combinan las técnicas de TOC-

IV con la de imagen por fluorescencia, dichos agentes de contraste permitirán imágenes

cardiovasculares de alta resolución.

En primer lugar, se ha investigado el nivel de contraste en las señales de TOC-IV dado por

diversas nanopartículas de oro, tanto a nivel de imágenes estáticas como a nivel de imágenes

dinámicas, utilizando las distintas funciones que la TOC-IV proporciona. Se ha determinado que

la TOC-IV es capaz de detectar nanopartículas de oro individuales, dada la gran sensibilidad de

ésta técnica de imagen.

En segundo lugar, dado que la TOC-IV se basa en la radiación esparcida por las diversas

muestras, se han investigado diversos tipos de nanopartículas de oro (nano-bastoncillos, nano-

estrellas y nano-esferas dobles (sílice-oro)) mediante diversas técnicas, incluyendo microscopía de

campo oscuro, espectroscopia foto acústica y, naturalmente, TOC-IV, concluyéndose que las nano-

esferas dobles (GNSs, de la traducción al inglés; “gold nano-shells”) son los mejores agentes de

contraste para TOC-IV.

En tercer lugar, se ha investigado el contraste producido por los GNSs tanto en células como en

tejidos, dadas las excelentes propiedades de estas nanopartículas. Para ello, se han incubado en

células HeLa cómo en células Jurkat, demostrando que se mejora el contraste en la señal de TOC-

IV dada por estas células una vez que se internalizan en ellas los GNSs. Este incremento de

contraste se ha verificado también mediante microscopía de campo oscuro. Además, se ha

demostrado que los GNSs insertados en tejidos incrementan el contraste en la señal de TOC-IV.

XVIII

Finalmente, se han investigado nanopartículas de cristales semiconductores (puntos cuánticos)

que emiten en la tercera venta biológica al ser excitados por el láser del sistema TOC-IV. Dada la

alta sensibilidad del TOC-IV, estos puntos cuánticos aumentan el contraste de la señal tomográfica,

de manera que actúan como dobles sondas ópticas, TOC y fluorescencia, tanto en sistemas

coloidales como en arterias y tejidos.

En consecuencia, las nanopartículas investigadas, tanto puntos cuánticos como nanopartículas

de oro, podrían ser muy útiles para imagen cardiovascular a nivel molecular, siempre que se

funcionalicen convenientemente para anclarse al sitio dañado. Esto proporciona altas posibilidades

de imagen con alta resolución en la formación de placas de aterosclerosis.

Palabras clave: Enfermedades cardiovasculares, placas de arterosclerosis, Tomografía óptica

coherente, nanopartículas de oro, puntos cuánticos

Chapter 1 Introduction and

motivation

Chapter 1 Introduction and motivation

2

1.1 Cardiovascular diseases

Cardiovascular diseases (CVDs) are the main cause of death in the worldwide, representing 32%

of all the death. The data was get from OECD. Stat,1 by selecting variables with causes of mortality

and diseases of the circulatory system, years are from 2012 to 2014. In China, the problem is more

serious, around 0.29 billion suffers, and a leading cause of death, which account for 42.61 % in

the urban area, according to the report on cardiovascular disease in China (2016). The main

dangerous factors for cardiovascular disease are hypertension, cigarette smoking, hyperlipidemia,

diabetes, insufficient physical activity, and air pollution.

Figure 1.1 The normal structure of the artery.2

Atherosclerosis is the key pathogenesis of cardiovascular diseases. It refers to the development

of atheromatous plaques in the arteries,2 notably in arterial branch points and bifurcations.3 The

normal structure of the artery consists of three layers. The adventitia layer, which is connected to

the tissue, contains mast cells, nerve ending and microvessels as shown in Figure 1.1.2 The tunica

media composed of smooth muscle cells (SMCs) embedded in a complex extracellular matrix. The

intimal layer, contains SMCs and an endothelial cell monolayer which exposures to the blood.2 In

healthy conditions, the arterial endothelial cells resist the adhesion of circulating immune cells to

them.4 Under atherogenic stimuli, such as hypertension, dyslipidaemia and cigarette smoking, the

changes in the endothelial permeability and the composition of the extracellular matrix in the

tunica media lead to entry and accumulation of low density lipoprotein (LDL) particles in the

subendothelial, as shown in Figure 1.2.5, 6 Once trapped in the subendothelium, some of the LDL

particles are oxidized, and thus cause the activation of the endothelial cells.7 The activated


3

endothelial cells express chemokines, that promote the recruitment of the monocytes to the vessel

wall and directional migration by interact with cognate chemokine receptors on the monocytes,

such as vascular cell adhesion molecule 1 (VCAM-1) and intercellular adhesion molecule 1

(ICAM-1). A number of monocytes in the early atherosclerotic lesions differentiate into

macrophages, and ingest oxidized lipoproteins via scavenger receptors.5, 8, 9 The accumulation of

lipid droplets in the cytoplasm of the macrophages, promote lipid-laden macrophages to become

foam cells, denote the atherosclerosis. Foam cells can secrete additional extracellular matrix

molecules. Extracellular matrix molecules are composed of certain collagen, elastin, proteoglycans

and glycoproteins. They interact with cells and generate signals to control the behaviors of the

cells.10 In atherosclerosis, they promote lipoprotein accumulation and recruitment of monocytes.8

Under prolonged endoplasmic reticulum (ER) stress, foam cells die from apoptosis and necrosis

and the cellular lipids are deposited in the artery. In the early atherosclerosis lesions, which is

called inflammation resolution, efferocytosis (the process that dying or dead cells are cleared by

phagocytic cells) is efficient and can rapidly clear the apoptotic macrophages. In advanced lesions,

so called no inflammation resolution, the efferocytosis is not efficient, the apoptotic macrophages

start undergo secondarily necrotic, and necrotic macrophages coalesce into necrotic cores. In the

lipid-rich necrotic cores, the death of the smooth muscle cells and degrading cap collagen would

thin the fibrous cap and promote the plaque vulnerable to rupture. In this process, two kind of

proteases: matrix metalloproteinases (MMP) and cysteine proteases play an important role in

arterial wall extracellular matrix degradation and remodeling.11-14 If there is no inflammation

resolution, the fibrous cap breaches leading to the luminal thrombosis and arterial occlusion.8

In the past few decades, molecular imaging has been developed to be a strong method to

visualize specific molecular process in vivo due to the fast development in imaging agents and

technologies. Molecular imaging can also make a significant contribution to the atherosclerosis

imaging. Normally, the application of molecular imaging to a special disease need to address the

following four question.15, 16 (1) Is there any molecular target corresponding to the disease of

interest? (2) Is there an affinity ligand that will bind to the target? Usually, the suitable molecular

target and ligand could be found in the literature. (3) What is the appropriate molecular imaging

system to provide the optimal resolution, sensitivity, and depth penetration for the disease? (4)

What kind of molecular agent is needed for the selected imaging system to detect the molecular

target?


4

Figure 1.2 The development of an atherosclerotic lesion, potential targets for molecular imaging at each

stage are listed on the top.6

For the first and second questions, different diseases and processes have different molecular

targets and ligands as shown in Figure 1.2.6 For example, in the early stages of atherosclerosis,

the endothelial cell dysfunction and activation caused by the hemodynamic and inflammatory

factors, lead to the expression of VCAM-1, ICAM-1 and selectins (P and E)7 on these

dysfunctional cells. The corresponding antibodies could act as affinity ligands, such as VCAM-1-

targeting peptides.17, 18 In the inflammation, the macrophages differentiated from monocytes can

be used as a target. Macrophages phagocytosis can be used as a target strategy to deliver the

nanoparticles (NPs). For example, their phagocytosis of LDL in the atherosclerosis process has

used the iron NPs conjugated with lectinlike oxidized LDL receptor-1 antibody to specifically

target to the macrophages.19 In the angiogenesis, the target is v3 integrin.

Molecular imaging studies of atherosclerosis are expected to be useful in the following clinical

scenarios:20 (1) identifying patients with high-risk of cardiovascular complications, (2)

characterizing the vulnerability of the lesions, (3) evaluating novel atherosclerotic therapies, and

(4) selecting individualized treatment strategies for the particular vulnerable plaques. In the


5

following section, I will talk about the imaging techniques and corresponding contrast agents, and

how they are used for the above clinical scenarios.

1.2 Cardiovascular imaging techniques based on light

1.2.1 Optical coherence tomography

1.2.1.1 Theory

Optical coherence tomography (OCT), which was discovered by David Huang and co-workers

in 1991, is very similar to ultrasonic pulse-echo imaging.21 It uses low coherence interferometry

to collect the scattering light from the tissue microstructures. OCT has very high resolution, around

15 µm and can detect backscattered light as small as 10-10 times of the incident intensity.21 Due to

these high resolution and sensitivity of the OCT, it has been widely used in medical areas, such as

ophthalmology and vascular surgery.22-25

In the particularly case of cardiovascular disease, the intravascular OCT (IV-OCT) was used. It

uses a single optical fiber (catheter) that can illuminate to the artery by rotating and being pulled

back along the artery,24 and at the same time, recording the backscattered light produced by the

artery. The reflected light then interferes with the light from reflected mirror at coupler, and then

is sent to the detector. The detector signal is converted to the reflectivity profile by using Fourier

analysis..

1.2.1.2 Applications of IV-OCT

Due to the high resolution of the OCT, it has been widely used to investigate the coronary

atherosclerosis. The normal structure of the artery as seen by IV-OCT, is shown in Figure 1.3 (a).

A healthy artery is formed by different layers, the intima (I), the internal elastic lamina (IEL), the

media (M), the external elastic lamina (EEL) and the adventitia (A). In the figure inside the artery

there is a catheter, with a diameter around 1 mm. In addition, the IV-OCT is also able to reveal the

morphology of atherma plaques, as shown in Figure 1.3. Therefore, it is possible to distinguish

different type of plaques, e.g. fibrous plaques with homogeneous and signal rich regions (Figure

1.3 (a)), fibrocalcific plaques with well delineated and signal poor regions with sharp borders

(Figure 1.3 (b)), and lipid rich plaques with signal poor regions and diffuse borders (Figure 1.3

(c)).26 IV-OCT is also useful for measuring vessel lumen diameter, which is very important for


6

selecting a suitable size of stent at a target site and helping the stent placement.27 In Figure 1.3 (a),

the diameter of the artery is around 2 mm. In Figure 1.3 (c), we see the fibrous cap, whose

thickness plays an important role in the plaque rupture. It has been observed that if the cap

thickness is less than 65 µm, the plaque can easy to break, and cause an acute thrombus.28

Therefore, OCT has also been used to measure the thickness of the fibrous cap as a marker of the

plaque vulnerability. A study about association of the statin therapy with reduced coronary plaque

rupture by OCT, showed the OCT measured thickness of the fibrous cap under stain therapy (78µm)

tend to increase than that of without stain therapy (49µm).29 In this case, after the stain therapy,

the coronary plaques tend to become more stable.

Figure 1.3 (a) OCT image of a fibrous coronary plaque showing a homogeneous, signal-rich interior (F).

An area of intimal hyperplasia is seen opposite fibrous lesion, demonstrating intima (I, with intimal

hyperplasia), internal elastic lamina (IEL), media (M), external elastic lamina (EEL), and adventitia (A).

(b) OCT image of a fibrocalcific aortic plaque showing a sharply delineated region with a signal-poor

interior. (c) OCT image of a lipid-rich carotid plaque showing a signal-poor lipid pool (L) with poorly

delineated borders beneath a thin homogeneous band, corresponding to fibrous cap (arrows).26 Bar=500µm

To avoid the attenuation that light suffers in the artery caused by the red blood cells, the blood

is completely removed with a saline flush before taking the OCT image.24 Xu and coworkers

probed that it is possible to identify the plaque composition from the IV-OCT signal by studying

the attenuation and the backscattering coefficient of different plaques. They found that the

backscattering coefficient had different values for calcification (4.9±1.5mm-1), for fibers

(18.4±6.4mm-1) and for lipid pool (28.1±8.9mm-1), and the same happened to the attenuation

coefficients (5.7±1.4mm-1, 6.4±1.2mm-1 and 13.7±4.5 mm-1, respectively).30


7

Other strategy for the characterization of atherosclerotic plaques is based on the study of

macrophages. Macrophages are inflammatory cells and they can release tissue factors and cause

plaque rupture.31 Some studies have shown that it is possible to detect the presence of macrophages

by OCT.32-34 As shown in Figure 1.4, which compares a fibrous cap with low density of

macrophages to one with high density of macrophages, the OCT intensity is much higher in the

case of high macrophages content (d). Macrophages can be identified in the low content

macrophage fibrous cap after doing the logarithm of the OCT image, and combining the images

with the ones given by histology.

Figure 1.4 Raw (a) and logarithm base 10 (b) OCT images of a fibroatheroma with a low density of

macrophages within the fibrous cap. (c), Corresponding histology for (a) and (b) (CD68 immunoperoxidase;

original magnification ×100). Raw (d) and logarithm base 10 (e) OCT images of a fibroatheroma with a

high density of macrophages within the fibrous cap. (f), Corresponding histology for (d) and (e) (CD68

immunoperoxidase; original magnification ×100).32

Currently, the IV-OCT is widely used for characterization of plaque type, fibrous cap thickness,

plaque rupture, and evaluation of intracoronary stenting.24, 35, 36 However, due to the lack of

molecular contrast, the molecular process under the atherosclerosis, as we discussed in the first


8

part, can not be visualized by IV-OCT. Therefore, the use of contrast agent would help IV-OCT

to detect the atherosclerosis process at molecular level.

1.2.2 Photoacoustic imaging

1.2.2.1 Theory

Photoacoustic (PA) imaging has become a strong tool in biological imaging due to its high

contrast, high resolution and high penetration depth. In PA imaging, the tissue is irradiated by a

short pulse laser with a duration of nano-second. The absorption of the light energy by the tissue

induces a slight temperature increment,37 which causes a thermoelastic expansion. This generates

a sound or stress wave (acoustic wave) that can propagate through the tissue. Once the acoustic

wave reaches to the tissue surface, it can be detected with an ultrasound transducer.38, 39

Normally, PA imaging is considered dependent on the absorption coefficient due to small

difference in the mechanical and thermodynamic properties of different tissues.39 The absorption

coefficient spectrum of endogenous tissue chromophores is shown in Figure 1.5. The differences

of the absorption coefficient of a single chromophore at different wavelength and chromophores

at the same wavelength allow the differentiation of the tissue type by properly choosing the PA

excitation wavelength.38 For example, by measuring the multiwavelength PA signal of the blood

vessel, the obtained concentration of oxyhaemoglobin and deoxyhaemoglobin can be used to

determine the blood oxygen saturation.40, 41

Figure 1.5 Absorption coefficient spectra of endogenous tissue chromophores.42


9

For PA imaging, penetration depth is dependent on the attenuations of the light and the acoustic

wave.42 In biological tissues, the light transfer is dominated by scattering.43 Therefore, the scattered

light can go through into deep of the tissue even they almost lose their original incidence

direction.43 However, these photons can still produce an acoustic wave, and this signal could still

be detect by the transducer because the acoustic scattering in the tissue is two or three order of the

magnitude weaker than that of light. Thus a high penetration depth for PA imaging is possible. For

example, the reported penetration depth achieved in human breast is 4 cm.44 The spatial resolution

is limited to sub-millimetric due to the acoustic attenuation, and also is determined by the width

of the pulse laser, width impulse response and focal diameter of the transducer, and the center

frequency of the received PA signal.37, 42

1.2.2.2 Applications to cardiovascular diseases

Due to its high penetration depth and not bad spatial resolution, PA imaging has been widely

used in biomedicine.37, 38, 42 In the case of cardiovascular disease, intravascular PA (IV-PA) has

been used for the identification of vulnerable plaque. In IV-PA, the axial resolution is 100µm,

lateral resolution is 500µm and the penetration depth is 5 mm.38 In the absorption spectra shown

in Figure 1.5, the absorption of lipid shows a strong absorption around 1200 nm. In 2010, by using

a 1200 nm laser as light source for IV-PA imaging, Wang et.al successfully got ex vivo image of

the lipid in atherosclerosis rabbit aorta based on the different optical absorption spectrum of the

fat in the near infrared range.45 Furthermore, a big number of contrast agents have been used to

get PA molecular images of the various stages of the atherosclerosis disease progression.46, 47 In

lipid rich necrotic cores, the elevated expression of matrix metalloproteinase-2 is one of the clinical

features. In 2016, Huan et.al used the GNRs conjugated with the matrix metalloproteinase-2

antibody as a high efficient IV-PA probe to detect the matrix metalloproteinase-2 in

atherosclerosis.48 The GNRs could target the matrix metalloproteinase-2 due to the antibody, and

thus increased the absorption of the matrix metalloproteinase-2, and a high IV-PA enhancement

was achieved which could provide a precise morphology to visualize the area of the distribution

of matrix metalloproteinase-2 in the arterial walls.48

IV-PA is currently undergoing refinement, although it has shown its capability of imaging the

atherosclerosis disease. Many important aspects face the real-time clinical imaging still need to be

solved, such as the optimal image acquisition sequence, laser source and catheter design.38


10

1.2.3 Fluorescence imaging

1.2.3.1 Fluorescence imaging technique

Fluorescence imaging is a technique that allows acquisition of data at high speed and

visualization of biological process in the dynamic conditions.49 Typically it uses light source (a

laser or white light with an excitation filter) to excite a fluorescent molecule, and the emitted

signals with different spectral characteristics can be isolated by emission filter and captured by a

CCD camera (detector range: 400 – 1000 nm) or an InGaAs camera (detector range 0.9-1.7 µm).50

This technique has two challenges. First is the limited penetration depth due to the absorption and

scattering of light due to the tissue. Therefore, it is very important to select the excitation and

emission wavelengths in the region known as biological windows which are shown in Figure 1.6.51

The first biological window (I-BW) extends from 700 to 950 nm, and corresponds to the end of

the visible absorption band from hemoglobin to the beginning of water absorption band around

980 nm.51 The second biological window (II-BW) is found from 1 to 1.35µm, in this window, the

scattering coefficient is reduced and a deeper penetration depth is expected in this range. The third

biological window (III-BW) is in the range from 1.5 to 1.8µm, the tissue autofluorescence is almost

zero in this range, which could allow getting a high optical contrast after using a contrast agent

that emits fluorescent signals in this range52. In second place is the need of appropriate contrast

agents. Differently from OCT and PA that can image the tissue based on the scattering and

absorption properties of the tissue without using exogenous agents, fluorescence imaging strongly

depends on contrast agent, due to the weak fluorescence given by most tissues.53 The contrast

agent has to be biologically stable in the organism, specially target the disease site, and produce

fluorescence imaging contrast.54 Common contrast agents emitting in BW have been widely used

for fluorescence imaging, such as ICG, cyanine dyes, quantum dots, single-walled carbon

nanotubes and rare earth nanoparticles.55


11

Figure 1.6 Absorption spectrum of human skin showing the first (I-BW), second (II-BW) and third (III-

BW) biological windows

1.2.3.2 Application to cardiovascular disease

During the last ten years, the huge progress in clinical intracoronary optical imaging lead to the

development of the near infrared spectroscopy catheter, which has been approved by the FDA.56,

57 Intravascular near infrared fluorescence (NIRF) imaging is a catheter based technique, in which

the excitation light is introduced to the artery by using a catheter, simultaneously, the emission

signal from the tissue is collected by the catheter and sent to the detector. Two dimensional

fluorescence images of the artery could be achieved by rotating and pull back the catheter. It has

been used to measure the disease activity within the cardiovascular.58 In 2011, Claudio et al

exploited the indocyanine green (ICG) as a contrast agent to imaging the atherosclerosis plaques.59

ICG is a FDA approved dye, could be quickly absorbed by lipid-rich plaques and cells. In addition,

ICG has a strong absorption in 785 nm, and emission at 815 nm, these absorption and emission

band are located in the I-BW. Therefore, ICG is expected to produce a high fluorescence contrast.

In this work, within 20 minutes injection, the sufficient contrast enhancement could be used in

vivo detection of lipid-rich, inflamed, coronarysized plaques in atherosclerotic rabbits. What’s

more, in vitro and human atheroma specimen studies showed that the ICG prefer to target the lipid

rich macrophages. At the same year, a dual-modality (OCT and NIRF) intra-arterial catheter for

simultaneous microstructural and molecular imaging in vivo was reported by Hongki et al.60 The


12

NIRF has a highly sensitive detection of Cyanine7 (Cy7), NIRF imaging of the fibrin-coated stent

in vitro and in vivo, thrombus and heterogeneous protease activity have been obtained by using

this technique.60

However, most of the contrast agent for intravascular NIRF imaging are dyes that are highly

susceptible to photobleaching, which will affect the sensitivity of the detection.53 Up to now, most

of the emission of these dyes are located in the first biological window, and the autofluorescence

of the tissue has a high effect on the NIRF signal. There, new contrast agent are strongly needed

for intravascular NIRF imaging.

1.3 Contrast agents

Nowadays, it is assumed that an efficient fight against CVDs requires a great effort in social

education for the introduction of healthy life habits and, in parallel, a significant improvement in

the diagnosis techniques. These techniques should improve in two distinct directions: Firstly, cost

reduction is mandatory, as convectional techniques (such as X-Ray assisted catheterization)

require expensive and complicate apparatus. Secondly, these conventional techniques fail, in most

of the cases, to provide molecular information of CVDs, so that the design and development of

specifically designed therapies are not possible. Nanotechnology can simultaneously overcome

these two challenges. The reduced size and high surface to volume ratio characteristic of

nanoparticles (NPs), makes them especially suitable for molecular imaging.61, 62 It has been already

demonstrated that NPs bound to specific affinity ligands, such as monoclonal antibodies, and so

enabling non-invasive phenotypic characterization of atherosclerosis plaques. At the same time,

the use of NPs also opens the door to the development of cost-effective and minimally invasive

diagnosis and imaging tools. In particular, the use of optical NPs as a contrast agents, i.e. NPs

capable of producing scattered light, luminescence or acoustic waves under light illumination,

would allow the use of cost effective laser light sources for cardiovascular imaging instead of more

complicated apparatus (such as MRI, X-Ray, SPET… and so on).61, 63-65 Because of these two

reasons, the scientific community is already considering optical NPs as reliable and powerful tools

for the next generation of imaging probes for the cardiovascular system.4, 66-68


13

1.3.1 Gold nanoparticles (GNPs)

Nowadays, GNPs have received a particular attention for bio applications, such as photothermal

therapy and medical imaging (OCT and PA). These widely applications of GNPs are due to the

following respects. First, they can be easily synthesized with different shapes and with tunable

optical properties. Second, they are biocompatible, can be functionalized with different ligand and

delivered to the disease sites.69 This part, I will focus on optical properties of GNPs and their

applications for cardiovascular disease.

Optical properties of GNPs: Surface plasmon resonance is the most important optical property

of the GNPs. It consists of a collective oscillation of conduction electrons induced by the

electromagnetic field of light. As shown in Figure 1.7 (a), after irradiated by light, the conduction

electrons oscillate coherently to the incident light. This results the absorption and scattering

properties of the GNPs. The oscillations wavelength strongly depends on the morphology of the

GNPs, for example, for GNRs, as shown in Figure 1.7 (b), they have two oscillations, transverse

and longitudinal, thus induce two extinction peaks in the extinction spectra.70 Furthermore, the

surrounding medium also affects the resonance wavelength.71

Figure 1.7 (a) Schematic drawing of the interaction of an electromagnetic radiation with a gold nanosphere.

A dipole is induced, which oscillates in phase with the electric field of the incoming light. (b) Transverse

and longitudinal oscillation of electrons in a gold nanorod.70


14

Gustav Mie solved the Maxwell’s equations for small spheres (d ≪ λ) obtaining an expression

that can predict the scattering and absorption properties of the particles.72, 73 The expression that

he obtained for the extinction coefficient Cext can be written as:

𝐶𝑒𝑥𝑡 =24𝑅3휀𝑚

3 2⁄

𝜆

휀2

(휀1 + 2휀𝑚)2 + 휀22 (1.1)

where R is the radius of the GNPs, εm is the medium dielectric constant, λ is the wavelength of the

incident light, ε1 and ε2 are respectively the real and imaginary part of the complex dielectric

function for small particles, and they are frequency dependent. Thus the extinction properties of

the small particles strongly depend on the dielectric constants of the metal, its surrounding medium

and its volume.51, 72 However, the Mie theory cannot be used to predict the optical properties of

nanoparticles with size larger than 20nm or explain the near infrared extinction spectra of

nanoparticles.51. In 1912, Gans developed a solution to explain the extinction properties of the

prolate spheroid. 74.The estimated extinction coefficient given by Gans74 is:

𝐶𝑒𝑥𝑡 =8𝜋2𝑅3휀𝑚

3 2⁄

3𝜆∑

(1 𝑃𝐽2⁄ )

{휀1 + [(1 − 𝑃𝑗) 𝑃𝑗⁄ ]휀𝑚}2

+ 휀22

𝑗

(1.2)

where Pj includes the depolarization factors for the each axis of the particle; Pa, Pb and Pc. For

gold nanorods (a > b = c), the depolarization factors are defined as

𝑃𝑎 =1 − 𝑟2

𝑟2[

1

2𝑟ln (

1 + 𝑟

1 − 𝑟) − 1] ; 𝑃𝑏 = 𝑃𝑐 =

1 − 𝑃𝑎

2 (1.3)

where r is corresponded to the aspect ratio (𝑟 = √1 − (𝑏 𝑎⁄ )2). The wavelength λmax longitudinal

plasmon resonance depend on the aspect ratio and medium dielectric constant has been derivated

by El-Sayed et al71:

𝜆𝑚𝑎𝑥 = 33.34휀𝑚

𝑏

𝑎− 46.31휀𝑚 + 472.31 (1.4)


15

For more complicated GNPs, such as gold nanoshells (GNSs) and gold nanostars (GNSTs),

some numerical methods have been developed to explain the absorption and scattering properties

of these GNPs. For example, discrete dipole approximation (DDA), finite difference time domain

and finite element method.75, 76

Figure 1.8. Tunability of the plasmon resonance maximum in nanoparticles. Variation of surface plasmon

extinction maximum λmax with (a) nanoshell total radius R2 at fixed R1/R2 = 0.857, (b) nanoshell core/shell

ratio R1/R2 at fixed R2 = 70 nm, (c) nanorod effective radius reff at fixed aspect ratio r = 3.9, and (d) nanorod

aspect ratio r at fixed reff = 11.43 nm (and straight line fit).77

Based on the methods that have been developed to explain the extinction spectra of the GNPs,

the extinction peak of the GNPs can be tuned to the near infrared region, in the biological windows,

which is very useful for applications in biology and medicine. For example, GNPs can act as a

contrast agent for bioimaging and be used for photothermal therapy72. For GNSs, the maximum

plasmon extinction λmax can be tuned from 800 to 1200 nm, when increasing the nanoshell total

radius R2 from 60 to 140 nm, while fixes the nanoshells core/total ratio R1/R2 at 0.857 as shown in

Figure 1.8 (a). In addition, this tunable property can be achieved by fixing the R2 at 70 nm and

increasing the ratio R1/R2 as shown in Figure 1.8 (b). In the case of the GNRs, the maximum


16

plasmon resonance can be shifted by changing the size (effective radius, which is given by 𝑟𝑒𝑓𝑓 =

(3𝑉 4𝜋⁄ )1 3⁄ ) and the aspect ratio (r) as shown in Figure 1.8 (c) and (d). Particularly, when fixes

the effective radius, the maximum surface plasmon extinction increases linearly as function of the

aspect ratio r.

Figure 1.9 Interaction between light and GNPs

As we already know, when the light excites the GNPs, it can be scattered and absorbed by the

GNPs, or pass through the GNPs, as illustrated in Figure 1.9. The absorption cross section and

scattering cross section determine the extinction cross section of the GNPs. Photothermal therapy

and PA application strongly depend on the absorption property of the GNPs.69, 78 However, OCT,

which detects the back scattered light of the material, is strongly dependent on the scattering

coefficient of the GNPs69. Therefore, the understanding of the absorption and scattering properties

of GNPs will make full use of the GNPs. In 2006, Prashant et al calculated the absorption and

scattering properties of GNPs of different sizes, shapes and compositions.77 As shown in Figure

1.10,77 for GNSs, when the core radius R1 is 120 nm, and the total radius R2 is 155 nm, the

scattering cross section start to play a big role in the extinction cross section after the extinction

wavelength at 600nm. They also gave a prediction to the scattering/absorption ratio (Csca/Cabs) of

GNPs based the morphology of the GNPs. They found that the Csca/Cabs ratio of GNSs increases

when increasing the total radius R2 while fixing the R1/R2 ratio. They also found that the ratio

increases when increasing the R1/R2 ratio while fixing the R2. As for gold nanorods (GNRs), the

aspect ratio r is the ratio of the nanorod dimension along the long axis to that along the short axis.

The Csca/Cabs of GNRs increases when increasing the effective radius reff, while fixing the aspect

ratio r. but it is very hard for Csca/Cabs ratio to increase to more than 1. The aspect ratio doesn’t


17

show a strong effect on the Csca/Cabs ratio when fixing the effective radius reff. Therefore, they

conclude that the absorption cross section is always the dominant factor in the extinction cross

section of the GNRs.

Figure 1.10 Tunability of the ratio of scattering to absorption of nanoparticles. Variation of Csca/Cabs with

(a) nanoshell total radius R2 at fixed R1/R2 = 0.857, (b) nanoshell core/shell ratio R1/R2 at fixed R2 = 70 nm,

(c) nanorod effective radius reff at fixed aspect ratio R = 3.9, and (d) nanorod aspect ratio r at fixed reff =

11.43 nm

Applications: Due to their strong absorption and scattering cross section and their strong

sensitivity to the environment, GNPs have been widely used for photo-thermal therapy, optical

contrast for OCT and PA, and biosensing.69, 78-80

In 2015, Adam de la Zerda et al used the GNRs for OCT imaging of living mice eyes, with an

OCT operating 860 ±48nm wavelength. The quantifying GNRs detection limit in mouse cornea

was achieved by injecting different concentration of GNRs (peaking at 780 nm) in mice corneas,

the limit detection concentration is 0.5 nM. The OCT contrast enhancement was achieved after

mice injected with 10 µL of GNRs at 50 nM indicated by the white arrow, compared to OCT cross

section image of mice only injected with balanced saline solution (BSS) and non-injected mice as


18

shown in Figure 1.11. The contrast enhancement is due to the scattering of GNRs. Therefore, they

demonstrated that GNRs could be used for OCT contrast enhancement.

Figure 1.11 OCT detection of GNRs in living mice corneas. Left: Optical eye images for mice corneas

were injected with 10 μL of GNR at 50 nM (lower), and control mice corneas were injected with 10 μL of

balanced saline solution (BSS) (middle) or not injected at all (upper). Right: OCT cross section images

through the mice eyes (left column, white dotted line).81

In addition to GNPs as a contrast agent based on the scattering properties, they also can be used

as a contrast agent based on the absorption properties. Gold nanocages, which has tunable

extinction properties has been studied as an absorption contrast agent for OCT by Younan Xia’s

group.82 As shown in Figure 1.12 (a), the gold nanocages show a high absorption coefficient,

being the absorption efficiency and cross section calculated by DDA 87% and 7.63 × 10-11 cm2,

respectively.82 The OCT image and signal distribution as function of depth showed a high OCT

signal attenuation in deeper depth compared to that of portion without gold nanocages, as shown

in Figure 1.8 (b) and (c). This is due to high absorption cross section and low scattering cross

section of gold nanocages.


19

Figure 1.12 (a) DDA calculated extinction, absorption and scattering coefficients of gold nanocages based

on the dimensions provided by the TEM image in the inset. (b) OCT image of a gelatin phantom embedded

with TiO2. The left portion did not contain any gold nanocages, while the right portion contained 1 nM of

gold nanocages. (c) Plots of the OCT signals on a log scale as a function of depth.

Furthermore, GNPs have been used as optical contrast agent for PA imaging, due to their

absorption cross section.46 In cardiovascular diseases, the matrix metalloproteinases play an

important role in weakening the plaque caps and promoting rupture.48 Therefore, the quantification

of the matrix metalloproteinases in atherosclerosis plaque is important in cardiovascular disease

treatment. Qin et al used GNRs conjugated with matrix metalloproteinases-2 antibodies for IV-PA

imaging detection of matrix metalloproteinases-2 in atherosclerosis plaque.48 The matrix

metalloproteinase-2 antibody can lead the GNRs specially target to the matrix metalloproteinases.

As shown in Figure 1.13, the IV-PA images show the structure of the aorta and the strong intensity

area is produced by the GNRs, which indicate the area containing matrix metalloproteinase. The

area of the bright PA signal region is almost the same as provided by the silver staining and

immunofluorescence, thus GNRs can be used as a promising probe for IV-PA imaging quantitative

detection of distribution of matrix metalloproteinases-2.48


20

Figure 1.13 (a) The photoacoustic image of the aorta containing plaque (b) Histological section of silver

stained (for labeling GNRs) aorta containing plaque from rabbits intravenously (ear) injected with GNRs

(1 mL, 1 mM) following 6 h circulation. (c) The immunofluorescence result of matrix metalloproteinases-

2 in the adjacent cross-sections. (d) Area of distribution of matrix metalloproteinases-2 measured using PA,

histological section of silver staining, and immunofluorescence methods, respectively.48

1.3.2 Quantum dots (QDs)

QDs are nanosizes semiconductor crystals that have been extensively used in fluorescence

imaging due to their small size, strong emission intensity and excellent photostability compared to

the molecular probes.55, 83-86 Their tunable optical properties obtained by changing their size make

them interesting for bioimaging, and their strong emission intensity could provide deep tissue

imaging. In this part, I will focus on the optical properties of the QDs and their applications for

cardiovascular imaging.

Optical properties: For a bulk crystal, the optical properties of the material only depend on the

chemical composition. However, when studying a nanosized crystal, the size of the particle starts

to change the optical properties of the material. Particularly, in small semiconductors, the

electronic structure changes from continuous bands to discrete or quantized electronic levels, due

to the apparition of electron-hole bound system. The energy of the electron and hole quantized

levels can be written as87


21

𝐸𝑙,𝑛𝑒,ℎ =

ℏ2𝜙𝑙.𝑛2

2𝑚𝑒,ℎ𝛼2 (1.5)

where ℏ is the reduced Planck constant, 𝜙𝑙.𝑛 are the root of the Bessel functions, which depend on

the principal quantum numbers n (1, 2, 3,…) and the angular momentum quantum number (0, 1,

2, 3…, corresponding to s, p, d,…, orbitals)88, me,h is the electron and hole effective mass

respectively, a is the crystal radius. Obviously, the energy of the electron and hole quantized level

decreases when increases the nanocrystal size.

Besides the quantization energy, the Coulomb interaction between the bounded electron and

hole also affects the optical properties of the QDs.89 The Coulomb energy is on the order of 𝑒2 휀𝑎⁄ ,

where ε is the semiconductor dielectric constant. Comparing the quantization energy and Coulomb

energy, they all increase with decreasing the radius of the semiconductor but in different order.

Therefore, in large QDs, the Coulomb interaction is much more important than the quantization

energy. On the contrary, when the radius of the QDs is very small, the quantization energy

produces main effect to the optical properties of the QDs.

Quantum confinement effects play an important role when the size of the nanocrystal reaches to

the natural length of the electron and hole, this is known as Bohr radius aB, which is calculated in

analogy to the hydrogen atom Bohr radius a0 (0.529nm). Thus aB is given as 89

𝑎𝐵 = 휀𝑚0

𝑚∗𝑎0 (1.6)

where m* is the effective mass, m0 is the mass of the electron. When the electron and hole are

confined together, they can form an exciton. The effective mass of the exciton is 𝑚𝑒𝑥𝑐∗ =

𝑚𝑒∗ × 𝑚ℎ

∗ (𝑚𝑒∗ + 𝑚ℎ

∗ )⁄ , 𝑚𝑒∗ and 𝑚ℎ

∗ is the effective of the electron and hole, respectively. After

introduced the effective mass of electron, hole and exciton in Equation (1.6), we can get the Bohr

radius for electron, hole and exciton respectively. For example, for the CdSe QDs, 𝑚𝑒∗ is 0.13 m0,

𝑚ℎ∗ is 0.45 m0, ε is 10.1, therefore, the exciton Bohr radius for CdSe QDs is 5.3 nm.90 In case of

PbS QDs, , 𝑚𝑒∗ is 0.087 m0, 𝑚ℎ

∗ is 0.083 m0, ε is 17, the electron, hole and exciton Bohr radius

(aB,e, aB,h and aB, exc) for PbS QDs is 10, 11 and 21 nm respectively.91, 92 The Bohr radius

determinates three kinds of quantum confinements regime.91 When a > aB, there is a weak

confinement regime. In the intermediate confinement regime, the QDs crystal radius is less than


22

the electron Bohr radius but larger than the hole Bohr radius (aB,h, < a < aB,e). Strong confinement

regime happens when the QDs crystal radius is less than Borh radius (a < aB), In this case, the

exciton energy is given by

ℏ𝜔𝜈 = 𝐸𝑔 + 𝐸𝑣ℎ(𝑎) + 𝐸𝑣

𝑒(𝑎) − 1.8𝑒2

휀𝑎 (1.7)

where Eg is the energy gap of the semiconductor and it also depends on the size of the QDs, the

Coulomb interaction is calculated by the first order perturbation theory87.

Figure 1.14 (a) Carton drawing and TEM a ZnS/CdSe QD. (b) Core size of ZnS/CdSe QDs and their

corresponding fluorescence imaging and emission spectra. (c) The corresponding energy gap and energy

level of the ZnS/CdS QDs corresponding to their core size.93

Figure 1.14 shows the optical properties of ZnS/CdSe QDs.93 In Figure 1.14 (b), when the size

of the core changes from 2.7 nm to 4.8 nm, the emission spectra show a red shift from 510 nm to

610 nm. Also in Figure 1.14 (c), the quantum confinement of the exciton lead to a decreasing

energy band gap when increases the core size. Therefore, by changing the size of the QDs, we can

get the emission we want for specific applications. In the case of the PbS, as shown in Figure 1.15,

the emission peak can be tuned from 752 to 1520 nm when increasing the size of PbS QDs from


23

2.35 to 5.61 nm.94 Thus the emission spectra of PbS can be tuned from I-BW to III-BW, which

will be very useful for different applications.

Figure 1.15 (a) Different size of PbS QDs. (b) Normalized absorption of PbS QDs. (c) Normalized

emission spectra of PbS QDs.94

Applications

Compared to fluorescent dyes, QDs have some advantages, such as long Stokes shift, broad

band excitation spectra, narrow emission bands and high photostability, thus they has been widely

used for high resolution cellular imaging, single particle tracking of biological process and ex and

in vivo imaging.95, 96 Furthermore, different QDs can be excited by single light source, therefore,

QDs can be used for simultaneously multicolor imaging.97

Particularly, QDs have been used for cardiovascular imaging. In 2016, Chenxin Wu and his

coworkers used ICG@PEG-Ag2S for atherosclerosis targeting and imaging in vivo.98 The Ag2S

QDs show a strong emission in NIR-II (peaking at 1200 nm).99 As shown in Figure 1.16, the

infrared fluorescence imaging of the aortas show two bright areas, which indicate the

atherosclerotic plaques and macrophages, these are due to the lipophilicity of the Ag2S QDs

surface to the atherosclerosis environment.98 The existence of the macrophages and atherosclerosis

in these areas were convinced by the Hematoxylin–eosin (H&E) and mouse F4/80 antibody

staining of the dissected aortas as shown in Figure 1.16 (b) and (c).


24

Figure 1.16 (a) NIR-II fluorescence imaging of the aortas, the aortas were harvested from the ApoE-/- mice

after intravenous injection of ICG@PEG-Ag2S. (b, c) Hematoxylin–eosin (H&E) and mouse F4/80

antibody staining of the dissected aortas corresponding to panel (a). White arrows indicate macrophages

and red arrows indicate atherosclerotic plaques.98

1.4 Targeting strategy

Targeting strategy is very important for improving the efficiency of NPs localizing to the disease

site and avoiding the fast clearance of NPs by the liver. Based on the properties of NPs and

different pathologies of disease, there are several targeting strategies that have been used to deliver

the NPs to the disease sites.

1.4.1 Passive targeting strategy

Enhanced vascular permeability: Enhanced vascular permeability is one of the main

mechanisms for passive targeting of diseases. This effect has been well investigated in cancer

therapy, it also exists in atherosclerosis due to the inflammation and rapidly formed of the leaky

microvessels in intima of the plaque. The vascular permeability allows NPs to penetrate the

vascular wall and accumulate in the inflammation sites.100 There are several factors that should be

taken into consideration, such as the size, shape and surface coating of the NPs,101, 102 and these

factors affect the clearance and circulation time of NPs in the vascular.100


25

Magnetically guided nanoparticles: The magnetic NPs could be used to image the

atherosclerosis plaque as they can be guided to the disease site by using an external magnetic

field.103, 104 This is the so called magnetically guided nanoparticles strategy. It is based on the

enhanced vascular permeability, and could improve the margination and accumulation of NPs in

the inflammation place.100

Shear induced target: For advanced atherosclerosis, one of the symptoms is the stenosis, which

results in the change of the wall shear stress. Take advantage of this effect, the shear stress sensitive

NPs can specially targeting to the plaques. Rui et al reported the use of up conversion NPs for

molecular imaging of vulnerable atherosclerosis.105 Based on the different wall shear stress in the

mouse, the NPs can specially target low shear stress site, where has vulnerable plaques. These

results have been confirmed by optical and MRI imaging, and the histological analyses.

1.4.2 Active targeting strategy

An scheme of the active targeting strategy has been given in Figure 1.2. The specially

functionalized NPs could be used to target specific disease sites. In this section, some examples of

using NPs to target endothelial cells and microphages have been discussed. More targeting

strategies related to different imaging techniques are shown in Table 1.1.

Endothelial cells: Due to their directly exposure to the bloodstream, endothelial cells has been

the main target of conjugated NPs for cardiovascular disease. Andrew et al studied the molecular

imaging with OCT by using microparticles conjugated with different antibodies.106 First they used

tumour necrosis factor (TNF-) to activate the human umbilical endothelial cells (HUVECs), this

step helps the expression of the VCAM-1, platelet endothelial cell adhesion molecule (PECAM-

1) and E-selectin on endothelial cells. The results show that under static conditions, NPs

conjugated with antibody bind to PECAM-1, E-selectin, VCAM-1 and E-selectin+VCAM-1.

Under flow conditions with increasing shear stress, PECAM-1-NPs bind great numbers than other

antibody-NPs. Furthermore, the binding of NPs to endothelial cells has been study in ex-vivo

coronary arterioles under different shear stress. The VCAM-1-NPs specially targeted to the TNF-

simulated arterioles. Due to the scattering properties of this NP, the binding of E-

selectin+VCAM-1-NPs to the simulated endothelial cells show a high OCT intensity than basal

without any binding NPs under different shear stress. Therefore, the OCT molecular imaging was


26

achieved by active targeting strategy.106 By using another technique, photoacoustic imaging,

Seunghan et al successfully detected and monitored the inflammatory response of HUVECs. After

simulated by TNF-, HUVECs could be expressed by cell adhesion molecules such as the ICAM-

1 and E-selectin. The antibody-GNRs were used to target these molecules, due to the specific

targeting and high absorption properties of these GNRs, a high PA signal was achieved for the

simulated HUVECs, Therefore, PA imaging could be used to monitor the inflammation response

by combining the active targeting strategy.

Table 1.1 Contrast agents (CAs) for cardiovascular imaging

CAs Imaging

techniques/conditions exc (nm) em(nm) Functionalization Targeting Disease Refs

Cy 5.5 In vivo fluorescence

imaging 550 680 Thrombin NA Thrombosis 107

Cy7 Ex vivo fluorescence

imaging 485 525 Fibrin NA Atherosclerosis 108

Cy5.5 In vivo and ex vivo

fluorescence imaging NA NA Profilin-I NA Atherosclerosis 109

NaGdF4:Yb,Er Visible fluorescence

imaging. 980 400-750 Osteopontin NA Atherosclerosis 105

ICG@Ag2S

Fluorescence imaging

(ex vivo) and PA (in vivo)

800 1200 C18/PEG NA Atherosclerosis 110

Gadolinium(III)–

GNRs

PA (In vitro, in vivo and ex vivo )

NA 520, 710 mPEG-SH Macrophage Atherosclerotic inflammation

111

GNRs PA (Ex vivo) 695 700 polyclonal matrix

metallopeptidase-2 antibody

Polyclonal

matrix

metallopeptidase-2

Atherosclerotic plaques

48

GNRs PA (In vitro) 700 700 anti-ICAM-1

Human umbilical

vein endothelial cells (HUVECs)

Inflammatory 112

GNSs PA (In vitro and in

vivo) NA 710 VCAM-1 antibody NA

Atherosclerotic plaque

113

GNPs IV-PA (Phantom and

ex vivo) 680 530 PEG Macrophage

Atherosclerotic plaques

114

GNRs PA (In vitro) 715;800 715;800 anti-ICAM-1

antibody and anti-E-selectin antibody

Human umbilical

vein endothelial cells (HUVECs)

Endothelial inflammation

115

PbS QDs OCT (Ex vivo) 1320 1550 DSPE-PEG-amine-

coated NA NA 116

Nanorose Photothermal OCT

(Ex vivo) 1328 NA Dextran Macrophage NA 117

Magnetic microspheres

Intravascular Magnetomotive-OCT

1310 nm NA RGD αvβ3 integrins atherosclerotic

plaque 118

Microparticles of iron oxide

OCT (In vitro and ex vivo)

1250-1350

NA Anti-VCAM-1, E-selectin, PECAM-1

HUVEC atherosclerotic

plaque 106

NA: Not available


27

Macrophages: Macrophages play an important role in the atherosclerosis progression because

they ingest the oxidized lipoproteins and lead to their differentiation into foam cells. Therefore,

macrophages imaging could be serve as biomarker of the atherosclerosis processes.119 High-

density lipoprotein HDL in macrophages helps the transport of the cholesterol to the liver for

excretion. Therefore, HDL-based contrast agents could be used to visualize macrophages. In 2013,

Marrache and co-workers developed (HDL)-mimicking NPs to detect macrophages apoptosis in

vitro. The core of these NPs contain QDs, which is used for optical imaging. The fluorescence

microscopy imaging showed the apoptotic macrophages by using these NPs compared to the

nontargeted NPs.120

1.5 Motivation

IV-OCT is a high resolution clinical technique to characterize the atherosclerosis plaque.

However, due to lack of contrast agent, the molecular processes under atherosclerosis progression

can not be visualized by IV-OCT. A strong demand of contrast agent to improve the IV-OCT

contrast has been called by the physicians. Based on the OCT technique, which collects the

backscattered light, NPs that could provide high backscattered light can be used to enhance the

OCT contrast. Therefore, the GNPs with high backscattered efficiency visualized by IV-OCT are

discussed in Chapter 3. The scatter properties determination of GNPs by photo-thermal method,

numerical simulation, dark field microscopy, PA are included in Chapter 4. Molecular OCT

imaging of the cells and tissue produced by GNSs was also achieved, the results and discussion

are given in Chapter 5. The fluorescence imaging of the atherosclerosis which is based on the

contrast agent could provide the molecular imaging of the atherosclerosis procession. Thus, the

combination of the fluorescence imaging and IV-OCT imaging could help to improve the accurate

treatment of the CVD. In this case, PbS QDs were evaluated as a contrast agent for fluorescence

imaging and IV-OCT in Chapter 6.

Chapter 2. Experimental

techniques

Chapter 2. Experimental techniques

30

2.1. Preparation of nanoparticles

2.1.1. Synthesis and functionalization of GNPs

GNSTs (surfactant-free) were synthesized by Dionysia Tsoutsi (Integrative Biomedical

Materials and Nanomedicine Lab, Department of Experimental and Health Sciences (DCEXS),

Pompeu Fabra University (UPF)). The synthesis process followed a seed-mediated growth

protocol, as previously reported by Yuan et al.121 The following chemicals were used:

Tetrachloroauric acid tetrahydrate (HAuCl4·4H2O), trisodium citrate dihydrate (C6H5O7Na·2H2O),

silver nitrate (AgNO3), L(+)-ascorbic acid, and hydrochloric acid (HCl), all of them purchased

from Sigma-Aldrich and used without further purification. Water was purified using a Milli-Q

system (Millipore). Aqua regia was used to clean all the glassware prior to use. Gold spheres, used

as seeds, were prepared by adding 15 mL of 1 wt % citrate solution to a 100 mL boiling solution

of HAuCl4 (1 mM) under stirring and additional boiling for 15 min. This is a standard citrate-

stabilized protocol producing wine-red colored gold sols with an average diameter of around 12

nm. Next, 1 mL of the above seed solution was added to a 100 mL of 0.25 mM HAuCl4 solution

containing 100 μL of 1 M HCl at room temperature and under moderate stirring. Immediately

afterwards, the addition of 1 mL of 0.01 M AgNO3 and 500 μL of 0.1 M ascorbic acid was

performed simultaneously and the solution was stirred for 30 s producing dark blue-black colloids.

The nanostar dispersions were centrifuged at 4500 rpm for 20 min, redispersed in MiliQ-water and

kept in the dark at 4 °C for long-term storage.

2.1.2. Synthesis and functionalization of QDs

Commercial CANdots® Series C (λem ~ 1600 nm) PbS QDs in a toluene dispersion (10 mg/mL,

1 mL) were added after careful sonication to CHCl3 (5 mL) while stirring at 23 ºC. In a second

flask 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-[amino(polyethylene glycol)-2000]

ammonium salt (DSPE-PEG-amine) (21 mg) was dissolved in CHCl3 (10 mL) and after complete

dissolution added dropwise to the first mixture. The combined mixtures were stirred for 2 h at 23

ºC and then the organic solvent was removed under reduced pressure. The obtained DSPE-PEG-

amine-coated QDs were then dispersed in distilled H2O


31

2.2. Basic characterization

2.2.1. Transmission electron microscopy (TEM)

TEM is a microscopy technique in which an electron beam is accelerated and focused in a

specimen (thickness <100 nm). The directions of the electrons change when they interact with the

atoms in the specimen, the scattered electrons thus produces an image. After magnified and

focused, the image can be seen on a fluorescence screen. The resolution of the TEM is higher 0.1

~ 0.2 nm, due to the de Broglie wavelength of the electrons.122 TEM is now widely used in the

chemistry, physics and biology.

The TEM we used in this work is a JEOL JEM 1010 microscope operating at 80 kV. Samples

were prepared by placing one drop of a diluted suspension of the nanoparticles in water onto a

carbon coated copper grid and leaving it to dry at room temperature. The size distribution of NPs

were determined by using ImageJ software.

2.2.2. Total reflection X-ray fluorescence (TXRF)

The concentration of PbS QDs was determined by using TXRF. TXRF uses X-ray to excite the

sample, which is put on the reflector, the angle between the incident beam and the reflector is less

than the 0.06°,123 thus the total incident photons are reflected. The X-ray fluorescence from the

sample is detected by putting a detector above the reflector. The detection limits is less than 10-12

g.124

The Pb-content was determined on a Bruker TXRF S2 PicoFox system, excitation by a Mo K

radiation. Calculations based on the size of the QDs and the PbS structure revealed the

concentration after ligand-addition as 0.6 mg/mL.

2.2.3. Extinction spectroscopy

The room temperature extinction spectrum of the colloidal solutions of nanoparticles were

measured with a double beam absorption spectrometer (Perkin Elmer Lambda1050).


32

2.2.4. Photoluminescence spectroscopy

The emission spectra of the PbS QDs was measured under excitation with a single mode fiber-

coupled 1280 nm laser diode. A long pass filter 1350 nm was put at the entrance slit of the Andor

Sahmrock 193i spectrometer to remove the excitation laser. After being analyzed by the

monochromator, the luminescence signal was collected by an InGaAs camera (Andor Idus

DU490A).

2.2.5. Dark field microscopy (DFM) imaging system

The DFM imaging system used in this thesis, shown in Figure 2.1 is composed of optical

microscope (NIKON Eclipse E400), dark field condenser (Nikon, dry NA 0.95-0.80), and different

objectives 50X (SLMPlan N, NA 0.35 WD 18 mm ), 40X (NA0.65, WD 0.65 mm), 20X (NA 0.40,

WD 3.8 mm), and 10X (NA 0.25, WD 6.1 mm). The signal is collected by InGaAs infrared camera

(top) and monochromatic (Shamrotic 193i). The visible image is detected by the EMCCD camera

(iXon Ultra Andor) which is connected to the monochromator, and the infrared emission is

detected by an iDus InGaAs detector (Andor). A filter wheel (Thorlabs) is located in front of the

infrared camera, in order to select the wavelength at which DFM are acquired. This DFM is

specially designed for experiment under flow conditions. A syringe pump is used to inject the

solution of GNPs in the flowing chamber. Meanwhile, a temperature controller (Warner, TC-324C)

is used to maintain the temperature in the flow chamber temperatures compatible with cell

aviability. The flow chamber (RC-30WA) is located on the platform bewteen the dark field

condenser and the objective.

The DFM working principle is shown in Figure 2.2 (a). A halogen lamp is used to illuminate

the specimen in the flowing chamber, and the central part of beam is blocked, so that only the light

scattered by the specimen can enter the objective. In our case, the specimen is located in the flow

chamber. The components of the chamber are shown in Figure 2.2 (b), which contains a pressure

plate on the top, a top coverslip plate, a top cover slip, a chamber forming gasket, a bottom cover

slip, a base, and a stage adapter. Vacumn grease is used to connect the coverslip to the top coverslip

plate and base. The temperature of the base is kept constant by the temperature controller (Figure

2.1). The cells are cultured on the bottom cover slip or top cover slip.The GNPs are introduced in

the flow chamber by the small tube in the top coverslip plate.


33

Figure 2.1 Dark field microscopy imaging system and flowing configuration (a) front and (b) back side.


34

Figure 2.2 (a) DFM schematic for flowing experiment. (b) RC-30WA flowing chamber

2.2.6. Photothermal method and experiments

Figure 2.3 (a) Schematic diagram of the experimental set-up used for the determination of the

scattering efficiency of GNPs under optical excitation in the II-BW. (b) Representative heating

and cooling curve obtained from an aqueous solution of GNRs under 1280 nm optical excitation.


35

The photo-thermal experiments were used to determine the absorption and scattering

efficiencies of the GNPs. The experimental setup is shown in Figure 2.3.

Briefly, a laser beam of power P0 at the required wavelength is directed onto a cuvette containing

a water solution with a certain concentration of GNPs (from now on the sample). A power meter

measures the output power P at the exit face of the sample. A fraction of the incident laser power

is attenuated by the GNPs depending on their scattering and absorption properties.125 Hence, the

transmitted power can be defined as follows:

𝑃 = 𝑃010−𝑂𝐷 (2.1)

where OD is the optical density of the sample that can be determined at different wavelengths from

the experimentally measured extinction spectra of each sample. We used three different continuous

wave (CW) laser diodes emitting at 808 nm, 980 nm, and 1280 nm. The fraction of laser power

absorbed by the sample will result in an increment in sample´s temperature that is measured by

using a thermal camera (FLIR E40bx) and record by FLIR tool program, as it is schematically

indicated in Figure 2.3 (a). The laser was kept on about 10 minutes (heating part of the Figure 2.3

(b)) until the temperature of the dispersion had reached a steady value. Then, it was turned off and

the temperature was monitored until it was equal to the room temperature, thus we obtained the

heating/cooling cycles, as the one shown in Figure 2.3 (b). The experimental determination of

𝜂𝑎(𝜆) requires the determination of temperature heating, (𝛥𝑇)𝑚𝑎𝑥 = 𝑇𝑚𝑎𝑥 − 𝑇𝑎𝑚𝑏 (where Tmax is

the maximum temperature and Tamb the ambient temperature) and the cooling decay time (𝜏). Both

magnitudes are schematically indicated in Figure 2.3 (b). The heat rate balance equation

accounting for these processes can be written as follows126:

𝑚𝐶𝑝

𝑑𝑇

𝑑𝑡= 𝑄𝐺𝑁𝑃 + 𝑄𝑟 − 𝑄𝑒𝑥𝑡 (2.2)

where m is the solvent mass, CP the constant pressure heat capacity, T the sample temperature and

QGNP the heat power delivered by the gold nanoparticles in the water solution. In Equation (2.2),

Qr is the heat power delivered by the solvent and the cuvette walls that can be measured for each

particular case, and Qext represents the external heat power losses, due to the temperature difference

ΔT between the sample and the surrounding environment. This term can be expressed as:


36

𝑄𝑒𝑥𝑡 = ℎ𝐴∆𝑇 (2.3)

where h is the heat-transfer coefficient and A is the cross sectional area for radiative heat transfer.

As we have assumed (no luminescence) that the heat power delivered due to the GNPs is equal to

the absorbed power, we can write:

𝑄𝐺𝑁𝑃 = 𝑃𝑎𝑏𝑠 = 𝑃0(1 − 10−𝑂𝐷)𝜂 (2.4)

When the laser is turned off (QGNP=Qr=0), Equations (2.2) and (2.3) lead to:

𝑑𝑡 = −𝑚𝐶𝑝𝑑𝑇

ℎ𝐴∆𝑇 (2.5)

whose solution is:

𝛥𝑇 = (𝛥𝑇)𝑚𝑎𝑥 ∙ 𝑒−𝑡𝜏⁄ (2.6)

where the experimentally measurable decay time constant, τ, is given by:

τ =𝑚𝐶𝑝

ℎ𝐴 (2.7)

To determine the heating efficiency we can rewrite Equation (2.1) under steady state conditions

(i.e. dT/dt=0) taking into account Equations (2.3) and (2.4):

𝜂𝑎 = 1 − 𝜂𝑠 =

𝑚𝐶𝑝∆𝑇𝑚𝑎𝑥

𝜏 − 𝑄𝑟

𝑃0(1 − 10−𝑂𝐷) (2.8)


37

2.2.7. Optical coherence tomography

Figure 2.4 (a) IV-OCT system. (b) IV-OCT catheter and rotary and pushing motor. (c) IV-OCT cross-

sectional image of artery.

For the experiments, the IV-OCT system used in this work is a commercially available system

for intravascular imaging ((Dragonfly™ OPTIS™ Imaging Catheter. St Jude Medical.)), which is

used for the treatment of patients at the Interventional Cardiology Unit University Hospital. Figure

2.4 show the real IV-OCT system in our lab, which is consist of computer system, rotary and

pushing motor and IV-OCT catheter. A cross-sectional image of artery is also presented in Figure

2.4 (c), in which we can see the IV-OCT catheter, the structure of the artery, a stent that is used to

expand the artery and metallic guide that is used to introduce the IV-OCT catheter. Figure 2.5 (a)

shows a schematic diagram of the OCT clinical system. A compact 1320 nm wavelength laser is

coupled to a single mode fiber (SMF) that is in turn incorporated into a 0.9 mm diameter catheter.

The SMF inside the catheter is optically coupled to a frequency domain interferometric system

that is constituted by different lenses and a rotating micro reflector that deviates the 1320 nm laser

beam 90º in respect to the SMF. The whole optical system is externally driven so that it can to

rotate at a constant angular frequency. As a consequence, as it is illustrated in Figure 2.4, the OCT

catheter is continuously scanning the 1320 nm beam over a cross sectional plane perpendicular to

it. The same optical system is used to collect the back-scattered 1320 nm signal and to couple it

into the SMF. The collected signal is then sent to a fixed-arms interferometer equipped with a low


38

loss spectrometer for frequency domain OCT. The nominal axial resolution of the clinical

intravascular OCT used in this work was close to 15 µm. In addition, the hydrophobic coated OCT

catheter is mechanically and optically coupled to a motorized unit, allowing for pull-back imaging

by retracting the catheter a total length of 5 cm at a speed of 20 mm/s. The scan diameter of the

OCT is 10 mm. The maximum frame rate and lines per frame achievable by our OCT system are

100 and 500, respectively. The emission spectrum of the OCT laser is included in Figure 2.5 (b).

The OCT spectrum consists of two lines, one at 1320 nm and another one at around 650 nm. This

second line arises from the fact that the OCT catheter incorporates a visible (red) laser that it is

used by clinicians to check if the catheter is fully operative before introducing it into the patient.

Figure 2.5 (a) Clinical IV-OCT set-up used all along in this work. (b) OCT laser spectra of the IV-OCT

The basic scheme of the low coherence interferometry (Michelson interferometry) used in OCT

is shown in Figure 2.6. The light source is a polychromatic plane wave with the electric field 𝐸𝑖 =

𝑠 (𝑘, 𝑤)𝑒𝑖(𝑘𝑧−𝑤𝑡) . Here, 𝑠 (𝑘, 𝑤) is the electric field amplitude, 𝑘 is the wavenumber and w is

angular frequency. The incident light is split by the beamsplitter into the sample and reference

arms into two equal electric field. After returning from the reference reflector and the sample, the

electric field for each arm is 𝐸𝑅 =𝐸𝑖

√2𝑟𝑅𝑒𝑖2𝑘𝑧𝑅 and 𝐸𝑆 =

𝐸𝑖

√2[𝑟𝑆(𝑧𝑆) ⊗ 𝑒𝑖2𝑘𝑧𝑆], respectively. Where

rR and rS(zS) are the electric field reflectivity of the reference reflector and samples, zR and zS are

the distance from the reference and sample arms to the beamsplitter, ⊗ denotes convolution, the


39

factor of 2 in the exponential is due to the path length in each arms. The returned light from the

sample arm is called backscattered light. It interferes after the beam splitter and the result of the

interference is detected by the detector, and produce a photocurrent, which can be expressed as

𝐼𝐷(𝑘, 𝑤) =𝜌

2⟨|𝐸𝑅 + 𝐸𝑆|2⟩. Here, ρ is sensitivity of the detector, the factor of 2 reflects the second

pass of each field through the beamsplitter.127 The received electric signal indicates the reflectivity

profile at different depth of the sample, this reflectivity profile is called A scan, which contains

information of spatial dimensions and location of structure. When the sample arm scans in lateral,

a cross-sectional tomography at each depth is achieved, this is B scan. Thus two dimensional cross-

sectional image of the sample is achieved by the OCT.

Figure 2.6 Schematic of Michelson interferometry used in OCT127

Normally, the electric field reflectivity rS(zS) in the biological tissue is continuous. Therefore

the rS(zS) could be written as 𝑟𝑆(𝑧𝑆) = ∑ 𝑟𝑆𝑛𝛿(𝑧𝑆 − 𝑧𝑆𝑛)𝑁𝑛=1 , where rSn corresponds to its electric

field reflectivity rS1, rS2…, and pathlength from the beamsplitter to sample zS1, zS2…. The power

reflectivity from each reflector in the sample and from reference reflector is given by the

magnitude squared of the electric field reflectivity,127 such as 𝑅𝑅 = |𝑟𝑅|2 and 𝑅𝑆1 = |𝑟𝑆1|2 ,

therefore, 𝑟𝑆 = √𝑅𝑆(𝑧𝑆). The rS is reintroduced into the electric field from sample arm, thus 𝐸𝑆 =

𝐸𝑖

√2[∑ 𝑟𝑆(𝑧𝑆)𝑒𝑖2𝑘𝑧𝑆𝑛𝑁

𝑛=1 ], and the detector current can be given:


40

𝐼𝐷(𝑘, 𝑤) =𝜌

2⟨|

𝑠(𝑘, 𝑤)

√2𝑟𝑅𝑒𝑖(2𝑘𝑧𝑅−𝑤𝑡) +

𝑠(𝑘, 𝑤)

√2∑ 𝑟𝑆𝑛𝑒𝑖(2𝑘𝑧𝑆𝑛−𝑤𝑡)

𝑁

𝑛=1

|

2

⟩ (2.9)

The angular frequency w is eliminated because the oscillation is so fast that the detector is not

able to respond. Therefore, by using the Euler’s rule: 𝑒𝑖𝑥 = cos 𝑥 + sin 𝑥, the detector current can

be written:127

𝐼𝐷(𝑘)

=𝜌

4[𝑆(𝑘)(𝑅𝑅 + 𝑅𝑆1 + 𝑅𝑆2 + ⋯ )] "𝐷𝐶 𝑡𝑒𝑟𝑚𝑠"

+𝜌

2[𝑆(𝑘) ∑ √𝑅𝑅𝑅𝑆𝑛(𝑐𝑜𝑠[2𝑘(𝑧𝑅 − 𝑧𝑆𝑛)])

𝑁

𝑛=1

] "𝐶𝑟𝑜𝑠𝑠 − 𝑐𝑜𝑟𝑟𝑒𝑙𝑎𝑡𝑖𝑜𝑛 𝑇𝑒𝑟𝑚𝑠"

+𝜌

4[𝑆(𝑘) ∑ √𝑅𝑆𝑛𝑅𝑆𝑚(𝑐𝑜𝑠[2𝑘(𝑧𝑆𝑛 − 𝑧𝑆𝑚)])

𝑁

𝑛≠𝑚=1

] "Auto

− correlation Terms".

(2.10)

Here, 𝑆(𝑘) = ⟨|𝑠 (𝑘, 𝑤)|2⟩, which represents the power spectral dependence of the light source.127

In Equation (2.10), the DC terms consists of the power spectral dependence of the light source

multiplied by the power reflectivity of the reference reflector plus the sum of the sample

reflectivities. It is an offset that is pathlength-independent to the detector current. The cross

correlation terms, also called the interferometric terms, strongly depend on the light source

wavenumber and the path length difference between the sample and reference arms. This is an

interesting part for OCT imaging and is less intense than the DC terms. The auto-correlation terms

is produced by the interference between the different sample reflectors. The best way to decrease

the auto-correlation effect is to select the proper reference reflectivity to increase the contribution

of the DC and interferometric terms in detector current. Therefore, a real detector current as

function of wavenumber is well explained.

There are two kinds of OCT system, time domain OCT (TDOCT) and Fourier domain OCT

(FDOCT). In TDOCT system, the light source is a broadband continous wave. The reference arm


41

consists of a mirror, which scans in length (zR) in order to reconstruct the sample reflectivity profile,

the reflected light from the reference and sample are recombined after pass the beamsplitter, and

their interference fringes are detected by a single channel photoreceiver. The detector current ID(zR)

in TDOCT can be obtained by the integration of Equation (2.10) over all k:

𝐼𝐷(z𝑅) =𝜌

4[𝑆0(𝑅𝑅 + 𝑅𝑆1 + 𝑅𝑆2 + ⋯ )] "DC terms"

+ρ

2[S0 ∑ √RRR

Sne-(zR-zSn)

2∆k

2

(cos[2k(zR-zSn)])

N

n=1

]

"𝐹𝑟𝑖𝑛𝑔𝑒 𝐵𝑢𝑟𝑠𝑡𝑠".

(2.11)

Here, 𝑆0 = ∫ 𝑆(𝑘)𝑑𝑘∞

0 is the integrated power of the light source. When the reference reflector

distance zR is equal to the sample distance zS1, for sample at position zS1, the detector current is

𝐼𝐷(z𝑅) =𝜌

4[𝑆0(𝑅𝑅 + 𝑅𝑆1)] +

ρ

2[S0√RRR

S1], as shown in Figure 2.7. In OCT system, a Gaussian-

shaped light source spectrum is used due to the Fourier transform properties. The power of light

source S(k) and its inverse Fourier transform γ(z) are

γ(z) = 𝑒−𝑧2∆𝑘2 𝐹↔ 𝑆(𝑘) =

1

∆𝑘√𝜋𝑒−[

(𝑘−𝑘0)

∆𝑘]

2

. (2.12)

Here, k0 (k0=2π/λ0) is the central wavenumber of the Gaussian shaped light source spectrum. Δk is

spectral bandwidth, which is equal to half-width of the spectrum at 1/e of its maximum intensity.

For the inverse Fourier transform γ(z), the full width at half the maximum (FWHM) is called

“coherence length”, lc. Based on (2.12), we use 1/2 to substitute γ(z), then 1/2 lc is √ln (2)/∆𝑘.The

coherence length in OCT is also called axial resolution, which can be expressed as

2√ln(2) 𝜆02/𝜋Δ𝜆.127, therefore, the coherence length lc is:

𝑙𝑐 =2√ln(2)

∆𝑘=

2√ln(2)

𝜋

𝜆02

Δ𝜆. (2.13)


42

Figure 2.7 A scan resulting from TDOCT127, it includes DC offset, and the fringe burst between the light

from the reference beam and sample beam at different distances (ZS1 and ZS2).

In the case of FDOCT, the reference beam is fixed, and the detector current is depend on the

wavenumber. A Fourier analysis is used to reconstruct the reflectivity profile. There are two kinds

of OCT based on the light source and detector. Spectral domain OCT (SDOCT), where a

broadband light source and detector array are used, which allows the simultaneously capture of all

spectral components of ID(k), and swept source OCT, also called optical frequency domain

imaging (OFDI), where a narrowband swept laser source is used. When the wavenumber of the

source is swept, the spectral component of ID(k) are captured in a single detector as function of

time.127

Given the Fourier transform relation 1

2[(𝛿(𝑧 + 𝑧0) + 𝛿(𝑧 − 𝑧0))]

𝐹↔ cos 𝑘𝑧0 and the

convolution property of Fourier transforms x(z) ⊗ y(z)𝐹↔ 𝑋(𝑘)𝑌(𝑘) , the inverse Fourier

transform of Equation (2.10) will be

𝑖𝐷(𝑘) =𝜌

4[𝛾(𝑧)(𝑅𝑅 + 𝑅𝑆1 + 𝑅𝑆2 + ⋯ )] "DC terms"

+𝜌

4[𝛾(𝑧) ⊗ ∑ √𝑅𝑅𝑅𝑆𝑛 (𝛿(𝑧 ± 2(𝑧𝑅 − 𝑧𝑆𝑛)))

𝑁

𝑛=1

]

"Cross-correlation Terms"

(2.14)


43

+𝜌

8[𝛾(𝑧) ⊗ ∑ √𝑅𝑆𝑛𝑅𝑆𝑚 (𝛿(𝑧 ± 2(𝑧𝑆𝑛 − 𝑧𝑆𝑚)))

𝑁

𝑛≠𝑚=1

]

"𝐴𝑢𝑡𝑜 − 𝑐𝑜𝑟𝑟𝑒𝑙𝑎𝑡𝑖𝑜𝑛 𝑇𝑒𝑟𝑚𝑠".

The resulting Fourier domain “A-scan” is shown in Figure 2.8.127 At the zero position, only DC

term appears in the detector current, this means the reflectivity profile appears at position of zR,

this is different from that of TDOCT, which appears at the beamsplitter. The distance different

between the reference reflector and the sample is doubled, this is due to the fact that the

interferometer measures the go and back distance to both reflectors.

Figure 2.8 A-scan resulting from FDOCT.127

Table 2.1 Comparison between TDOCT and FDOCT

Specifications TDOCT FDOCT

Axial scans/s 5,000–10,000 ～100,000

Lines/frame ～200 ～500–1,000

Max. frame rate, fps 20 ～200

Max. pullback speed, mm/s 2 20

Axial resolution, µm 15 10-15

Lateral resolution, µm 90 20-40

Tissue penetration, mm 1.5-3 2-3.5

Let’s see the differences between TDOCT and FDOCT, the specifications of the TDOCT and

FDOCT are shown in Table 2.1.24 First, both of them show a high axial resolution than that of

intravascular ultrasound (IVUS), in IVUS, the axial resolution is around 100µm.128 Second, the


44

FDOCT system is faster than TDOCT, which is due to the fixed mirror in the reflector arms and

the variable frequency light source or swept laser in FDOCT system. The IV-OCT is developed

based on the FDOCT.

IV-OCT cross-sectional image analyze

Integrated OCT intensity: The IV-OCT video was imported in the ImageJ software, then an area

without catheter , but inside the tubing was selected in the IV-OCT cross sectional image, we

added this selected area in the ROI, and used the multi measure option to measure the intensity.

The integrated intensity, intensity per pixel, maximum and minimum intensity could be obtained

at the same time.

Individual spot analysis: Here we show you an example of how we analyzed the HeLa cells

incubated with GNSs. The IV-OCT video was imported in the ImageJ software, first we need

select the area where has GNSs and cells (inside the tubing) by change the contrast of the tubing

and the catheter to zero (Brightness /Contrast option, adjust the minimum from 0 to 255). Then we

change the image type to 8-bit. After this, we adjust the threshold from 0-255 to 37-355, because

the threshold is the OCT intensity, and the intensity of the background is around 37, therefore, we

adjust the threshold to remove the effect of the background. The statistics are carried out by using

the analyze particles option, the size (pixel^2) is set to 3-30, as the size of the GNSs and cells are

large than 1 pixel. Select the display results, and press ok, you will have the number of spots and

max intensity per spot. Save it, import it to origin and do the frequency counts, you will have the

number of spots for each spot intensity.

2.2.8. Infrared fluorescence imaging

For fluorescence imaging, the IV-OCT laser was used as an excitation source, a XEVA-320

infrared camera was used to take the infrared fluorescence imaging. This camera is based on an

InGaAs two-dimensional array cooled down to 0 ºC that makes the detection of fluorescence

images in the 900-1700 nm spectral range possible. A long pass filter 1350 nm was used to remove

the effect from IV-OCT laser.


45

2.2.9. Photoacoustic experiment

The photoacoustic experiment was carried out in Optoelectronics and Laser Technology Group

(GOTL), Universidad Carlos III de Madrid. With the permission from Prof. Horacio Lamela

Rivera, and help by Luca Leggio and Daniel Gallego.

The photoacoustic signals were obtained by using a fiber-coupled Nd:YAG laser operating at

1064 nm and providing 8 ns width pulses as excitation source. The laser had a repetition rate of 1

kHz and a pulse energy of 10.5 µJ, leading to an average power of 10.5 mW. The GNPs were

placed in a cuvette with thickness of 4 mm. In order to perform comparable measurements, the

concentration of GNPs in the solutions was adjusted to provide an optical density (O.D.) of 10 at

the laser wavelength (1064 nm). The photoacousti signals generated in the sample were detected

by an ultrasound transducer mechanically attached to the cuvette. The signal generated by the

ultrasound transducer was sent to a 40 dB pre-amplifier and collected by a digital oscilloscope.129

Basically, PA is also called optoacoustic or thermoacoustic imaging. In optoacoustic imaging,

the photoacoustic signal is induced by the light, while in thermoacoustic imaging, the PA signal is

produced by the radiofrequency.37 For intravascular photoacoustic imaging (IV-PA), the light is

delivered by an optic fiber and the signal is collected by the transducer in the same optic fiber.

This technique is used for imaging the plaque structure and its composition in atherosclerosis.

An image of the tissue can be achieved by scanning the ultrasound.38 The PA source strength p0

can be estimated by

𝑝0 = (𝛽𝑐2/𝐶𝑝)𝜇𝑎𝐹 = ΓA, (2.15)

where is the isobaric volume expansion coefficient, c is the speed of the sound in vacuum, Cp is

the specific heat at constant pressure of the sample, µ is the absorption coefficient of the sample,

F is local energy density, is the Grüneisen parameter: Γ = 𝛽𝑐2/𝐶𝑝 , A is the local energy

deposition density: A = 𝜇𝑎𝐹.37, 38


46

2.3. In vitro experiment

2.3.1. HeLa and Jurkat cells; culture and incubation with Gold Nanoshells (GNSs)

Two different cell lines were investigated, i.e., HeLa cells and Jurkat cells. HeLa cells are an

immortal cell line derived from human cervical cancer cells, with an epithelial phenotype. These

cells divide an unlimited number of times in vitro under minimal survival conditions; accordingly,

they are commonly used for cancer research130. In addition, as is typical for cancer cells, they are

able to take up nanoparticles by passive targeting, usually by endocytosis. HeLa cells (ATCC

CCL-2) were grown in Dulbecco’s modified Eagle’s medium with 10% fetal bovine serum (FBS).

Jurkat cells are an immortalized line of human T lymphocytes. They (ATCC TIB 152) were grown

in RPMI 1640 medium with 10% FBS. Both cell lines were cultured at 37 °C under 5% CO2 and

95% humidity. Cells in exponential growth were used throughout all experiments. Incubation

experiments were performed at 37 °C for 24 h using PBS solutions containing various

concentrations of GNSs (0.05, 0.1, 0.5, 2.5, and 5 µg/mL). After incubation, cells were washed

several times to remove excess GNSs that did not adhere or were not internalized by the cells.

After incubation with GNSs, HeLa cells were trypsinized. The detached cells were resuspended in

PBS and centrifuged at 1,000 rpm at 4 °C for 5 min. This step was repeated three times to eliminate

excess NPs. Jurkat cells were resuspended without trypsinization in PBS and subjected to three

cycles of centrifugation in order to eliminate excess GNSs.

2.3.2. Human Mammary Epithelial Cells culture, activation and evaluation of molecular

expression

Human microvascular endothelial cell line (HMEC-1) was obtained from the American Type

Culture Collection (Teddington, UK; ATCC, No CRL-3243) and was grown in 5% CO2 at 37 °C

in a medium containing endothelial basal medium MCDB 131 (Gibco), 10% heat-inactivated FBS,

1% L-glutamine, 100 μg/ml each penicillin and streptomycin, 100 nM hydrocortisone (Corning),

and 10 ng/ml human epidermal growth supplement (EGF; Gibco BRL). For flow cytometry assay,

the cells were cultured on 6-well plates (FALCON, USA). Only cells with less than 12 passages

were used in the present study.


47

Figure 2.9 Flow cytometer analysis of the expression of ICAM-1, ICAM-1 Sb and VCAM-1 in (and

without) TNF--activated HMEC-1. Data in histograms are also displayed in bar graphs in the bottom of

this figure.

HMEC-1 (200,000 cells/well) in six-well plates (Costar, Cambridge, MA, USA) were

stimulated with 40 ng/mL TNF- (DakoCytomation, Glostrup, Denmark) for 18 h. To evaluate

the expression of ICAM-1 or VCAM-1 after exposure to TNF-, three kinds of molecules were


48

used for the expression, ICAM-1 purified, ICAM-1 supernatant of hybridome (Sb) and VCAM-1,

the cells were detached with a solution of 0.05% trypsin-EDTA, which was afterwards neutralized

by the addition of MCDB131 complete medium. Then, the cells were washed in ice-cold PBS by

centrifugation (5 min at 4°C) and resuspended in PBS containing mouse monoclonal anti-human

VCAM-1/CD106 (MCA907. BIORAD) or mouse anti-human ICAM-1 monoclonal antibody

(TONBO, San Diego. CA). After incubation for 1 h at 4°C, the cells were washed with ice-cold

1% BSA-PBS to remove the unbound antibody. The cell pellet was recovered by centrifugation (5

min, at 4°C) and resuspended in cold PBS containing FITC-conjugated rabbit anti-mouse Ig (1/40)

(Dako, Den- mark). After washing, the cells were resuspended in PBS before being analyzed by

flow cytometry with a FACScan flow cytometer (model FACS- Calibur, Becton Dickinson, San

Jose, CA, USA) with the CellQuest analysis program (BD Biosciences, San Diego, CA, USA).

Flow cytometry was performed using the appropriate single-stained samples for compensation

setting. The number of cells expressing adhesion molecules was determined according to a forward

light scatter/side light scatter gating combined with an FL-1 channel for immunostaining. The

results are shown in Figure 2.9, before the HEMC-1 activated by TNF-, the expression of ICAM-

1 and ICAM-1 Sb are 80% and 70%, respectively. After activated by TNF-, the expression of

these two molecules have reached to 100%. In case of VCAM-1, the expression efficiency is very

low, only 10% after TNF- activated.

2.3.3. Cell viability study

In order to evaluate the cellular toxicity of GNSs used in the my thesis, the MTT (3-(4,5-

dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) tetrazolium reduction assay was used in

all the cases. Metabolically active cells are capable of converting MTT to formazan, which is a

water-insoluble compound with dark blue color. While the dead cells lose the ability to produce

formazan from MTT. When the formazan is formed in culture medium, they precipitate inside

cells and being deposited inside the culture medium due to their insoluble properties in water. Here,

Dimethylsulfoxide (DMSO), an organic medium, is used to stabilize the formazan and avoid

evaporation.131 Thus formazan can be quantified by from the optical absorption of the dispersion

at 540 nm, and finally the number of living cells are obtained compared to the control.


49

In this thesis, the cell viability was studied on HeLa cells. To carried out these test, the cells

were cultured in a plate with 24 wells at 37 °C in controlled atmosphere (5% of CO2). After 48

hours, the culture medium was replaced by new culture medium in which NPs had previously been

dispersed at different concentrations. The HeLa cells were incubated with GNSs for 24 hours. At

the end of the incubation period, the culture medium with NPs was removed and medium without

NPs was added to these wells. After 24 hours, 0.5 mL of an 0.05 mg/mL MTT dispersion was

added to each wells and incubation was continued in controlled atmosphere for 3 hours, after that,

the culture medium were discarded. The formazan crystals were dispersed in 0.5 mL of DMSO.

Finally, the optical density of the samples at 540 nm was measured using a reader of microplates

(Espectra Fluor 4, Tecan). Based on it, the percentage of feasibility of cells incubated with NPs

with respect to the control case was calculated by using the following expression:

Viability (%) =𝐴𝑁𝑃𝑠

𝐴𝑐𝑜𝑛𝑡𝑟𝑜𝑙× 100 (2.16)

Where ANPs is the absorption of formazan produced by cells after incubated with NPs, and

Acontrol is the absorption of formazan produced by cells without incubation.

Chapter 3. Gold nanoparticles

in water dispersion

visualization by OCT

Chapter 3. Gold nanoparticles in water dispersion visualization by OCT

52

3.1. Introduction

Although IV-OCT systems have already demonstrated to be an outstanding for both diagnosis

and assistance for stent implantation, their unsurpassed sensitivity, resolution and speed suggest

that they are also suitable for application in other fields such as intravascular flow studies or

imaging of vessel walls at the molecular scale.60 Nevertheless, such novel applications would

require the use of optical probes capable of providing additional optical contrast at the operating

OCT wavelength of 1.3 µm. At the present time few OCT contrast agents have been already

developed, including gas-filled microbubbles and micrometer-sized droplets of oil with embedded

silica particles on their interface.132, 133 However, their clinical application is somehow limited due

to their micrometric size. Such limitation can be easily overcome by using nano-sized OCT

contrast agents, in particular by using gold nanoparticles (GNPs).

GNPs have shown to be excellent contrast agents for a variety of imaging techniques.69, 134 In

addition, they display a number of advantages for biomedical applications.135 (i) They are inert,

biocompatible 136 and can be properly functionalized with specific biological markers, allowing

specific targeting and making possible the so-called “personalized diagnosis”. (ii) GNPs can been

easily synthesized by a variety of methods leading to different shapes (spheres, rod, cages, prisms,

and stars) and sizes,72, 77, 137 as well as combined core (dielectric)/shell (gold) spherically

nanostructures, usually called gold nanoshells (GNSs). (iii) The morphology of GNPs determines

the spectral position of their localized surface plasmon resonance (LSPR), which is related to the

collective oscillations of free electrons confined at the dimensions of such nanometric particles.

By tuning GNPs morphology, their optical response can be custom-made depending on the

application.78 Indeed, it is possible to shift the LSPR of GNPs from the visible to the BWs (700-

950 nm, 1-1.3 µm, 1.5-1.8 µm), to enable in vivo applications. (iv) The scattering and absorption

properties of GNPs can also be tuned depending on the selected NP’s geometry.138, 139 Thus,

enabling the use of GNPs as photo-thermal transducers for in in-vivo treatments, as contrast agents

for optical imaging (multiphoton excited fluorescence imaging and scattering based imaging

techniques), for PA imaging and for X-Ray Computed Tomography.46, 51, 69, 80, 140-142

In fact, nowadays, it is possible to find out numerous examples on the use of GNPs to increase

contrast in OCT images. In most of cases non-clinical OCT equipment working at wavelengths


53

close to 800 nm were used, so that gold nanorods (GNRs) with LSPR 800 nm were used as

contrast enhancers.143 Despite the variety of results published dealing with GNPs as contrast agents

for OCT imaging, these nanoparticles have not been evaluated yet as contrast enhancers in OCT

cardiovascular imaging.

In this chapter, a systematic investigation on the potential use of GNPs as contrast agents for 1.3

µm wavelength based clinical IV-OCT systems for cardiovascular imaging was conducted. The

importance of the spectral overlap between the OCT laser and the GNP extinction spectra is

investigated by performing experiments with GNRs and GNSs water solution displaying LSPR

ranging from 800 up to 1300 nm. The function of IV-OCT system has been investigated by static

and pull back operation after the injection of GNSs inside the tubing.

3.2. Characterization of GNPs

I will focus on the study of different GNPs with surface plasmon resonance wavelengths in the

NIR, not far from 1.3 µm, as this is the operating wavelength of our IV-OCT system. Two types

of commercial GNPs showing plasmonic extinction in this wavelength were used: GNRs and

GNSs. GNRs were purchased from Nanopartz Inc., whereas GNSs were purchased from

nanoComposix Inc. Figure 3.1(a) includes representative TEM images of the different GNPs

obtained from these two companies. Table 3.1 summarizes their main characteristics, including

dimensions, nominal localized surface plasmon resonance wavelength (LSPR), effective radius,

and mass per GNP. As can be observed, up to a total of six different GNPs were studied; four

GNRs with nominal LSPR wavelengths peak at 800, 1064, 1200 and 1300 nm and two GNSs with

resonances peak at 856 and 986 nm. The extinction cross section spectra of the different GNPs

used in the thesis are shown in Figure 3.1(b) for GNRs and Figure 3.1(c) for GNSs. The emission

spectrum of the IV-OCT laser is included in both cases for the sake of comparison. The extinction

cross section at the particular IV-OCT operating wavelength (ext (OCT)) for each type of GNP

has also been included in Table 3.1. From the Figure 3.1(b), it is clear that GNRs show relevant

extinction cross sections not only at the plasmon resonance wavelength (at which extinction cross

section peaks) but over a quite broad spectral range. There is a shoulder peaking at around 500 nm

for these GNRs, this is due to the transverse plasmon resonance, the other one correspond to the

longitudinal plasmon resonance.71 The spectral broadening of the extinction cross section spectra


54

is mainly attributed to the existence of GNRs of slightly different sizes and shapes in the water

dispersion (polydispersity). As the morphology (size and aspect ratio) of GNRs determine the

spectral location of their LSPR wavelength, such polydispersity leads to an inhomogeneous line

broadening in the extinction spectra.77, 144, 145 As a consequence of this spectral broadening, all

GNPs present a non-vanishing extinction cross section at the OCT wavelength (𝜎𝑒𝑥𝑡(𝜆𝑂𝐶𝑇), see

detailed values in Table 3.1). Note that all the GNRs here investigated showed extinctions cross

sections at OCT in the order of 10-12 cm2 per nanoparticle. In the case of GNSs, as shown in Figure

3.1 (c), much broader spectra are obtained in comparison to GNRs. Such broad extinction spectra

are expected for GNSs due to the optical activation of surface charge oscillations of different

orders.77, 146, 147 Due to this superior broadening, although both GNSs have their LSPR below 1000

nm, they show non-vanishing extinction cross section values at OCT, more than two orders of

magnitude larger than those obtained for GNRs, being in the order of 10-10 cm2 per nanoparticle

for both GNSs.


55

Figure 3.1 (a) TEM images of the different gold nanoparticles investigated in the thesis (b) and (c) show

the extinction cross section spectra obtained for the GNRs and GNSs, respectively. The OCT laser spectrum

is displayed in (b) and (c) as a black line.

Table 3.1 Optical and physical properties of the different types of GNPs studied in the thesis. The

OCT intensity (dB) per nanoparticle is also included.

GNP NSPR

(nm)

Dimensions

(nm)

reff

(nm)

Mass/GNP

(gr/GNP)

ext @OCT

(cm2) abs

sct @OCT

(cm2) dB/GNP

GNR 800 800 25×90 21.9 7.72×10-16 3.66×10-12 0.66 1.24×10-12 9.80×10-11

GNR 1064 1064 10×67 10.8 9.80×10-17 3.24×10-12 0.96 0.12×10-12 1.71×10-11

GNR 1200 1200 21×171 24.2 7.84×10-17 1.83×10-12 0.76 0.44×10-12 2.30×10-11

GNR 1300 1300 10×100 12.3 1.47×10-16 5.31×10-12 0.96 0.20×10-12 3.00×10-11

GNS 856 856 15.5×117.8 74.1 1.67×10-14 3.61×10-10 0.34 2.40×10-10 3.96×10-9

GNS 986 986 16.4×198.1 115.5 6.80×10-14 7.63×10-10 0.37 4.84×10-10 1.15×10-8

Previous works pointed out the relevant role played by the scattering cross section (an in

particular by the backscattering cross section) of nanoparticles in their OCT enhancement

capability;148, 149 the higher value of backscattering coefficient the higher OCT intensity gain. Thus,

in order to make a proper interpretation of OCT images, the scattering cross section at OCT

(𝜎𝑠𝑐𝑡(𝜆𝑂𝐶𝑇)) for each type of GNP has been estimated. These cross section values have been

determined from the experimental value of the extinction cross section and the particular

absorption efficiency of each type of GNP. The absorption efficiency of any GNP is defined as the

ratio between the absorption and extinction cross section, i.e. 𝜙𝑎𝑏𝑠 =𝜎𝑎𝑏𝑠(𝜆𝑂𝐶𝑇)

𝜎𝑒𝑥𝑡(𝜆𝑂𝐶𝑇).51, 150 This

efficiency can be estimated by using the discrete dipole approximation method and, in particular,

the open-source Fortran-90 software package Discrete Dipole Scattering (DDSCAT) 7.3.76 Once

𝜙𝑎𝑏𝑠 is known, the scattering cross section can be estimated from the extinction cross section by

taking into account that 𝜎𝑒𝑥𝑡 = 𝜎𝑎𝑏𝑠 + 𝜎𝑠𝑐𝑡. Thus, we can write:

𝜎𝑠𝑐𝑡(𝜆𝑂𝐶𝑇) = (1 − 𝜙𝑎𝑏𝑠)𝜎𝑒𝑥𝑡(𝜆𝑂𝐶𝑇) (3.1)


56

Table 3.1 includes both the absorption efficiencies and the estimated scattering cross sections

at OCT operating wavelength for all the GNPs under study. It is clear that GNSs display the largest

values of 𝜎𝑠𝑐𝑡(𝜆𝑂𝐶𝑇) among all the studied GNPs, this fact is due to their larger extinction cross

section values (when compared to GNRs) as well as to their reduced absorption efficiencies (i.e.

to their large scattering efficiencies).

Previous works clearly stated that the OCT contrast enhancement provided by nanoparticles

basically depends on the relative change in the backscattering coefficient between the nanoparticle

and the medium.143, 151 For instance, in highly scattering media, nanoparticles with low scattering

cross sections have demonstrated to lead to a maximum OCT contrast.143 In this case, a weakly

scattering medium was used, and so, the achievement of high OCT contrast would require the use

of nanoparticles with high scattering cross section and, in particular, with high back-scattering

cross section, σbsct.21 In this respect, the DDSCAT code allows also for estimating σbsct for any

GNPs. The estimated values using this code are also listed in Table 3.1. GNSs are expected to be

those providing the largest OCT contrast over all the GNPs here considered. In order to corroborate

this prediction, the OCT contrast generated per individual GNP was systematically investigated

for all the GNPs isotonic saline solution.

3.3. Static OCT imaging experiments

The experimental procedure followed all along the chapter for the visualization of GNPs is

schematically represented in Figure 3.2. Basically, the IV-OCT catheter is introduced in a plastic

tubing of 3 and 3.6 mm of internal and external diameter, respectively. GNPs were then injected

in the proximity of the OCT catheter by using a subcutaneous needle (Figure 3.2 (a)). A volume

of 5 µL of a solution of GNPs in isotonic saline solution was injected. Solutions containing

different concentrations of GNPs were used. After the injection, three different types of

experiments for GNP visualization were performed, schematically drawn in Figure 3.2 (b-d). A

representative IV-OCT slice provided by IV-OCT system was also included in Figure 3.2 (e). It

shows two cross-sectional images, one is the cross-sectional image of the tubing filled with GNPs,

the other one is the axial cross-sectional image obtained from the line crossed the tubing and

catheter. More information, such as the IV-OCT scan time, the date of the experiment, the

morphology of the tubing and the table where the tubing was put can be obtained from this image.


57

Particularly, I will focus on the OCT cross-sectional image contained in the dotted line circle,

which contains a cross-sectional image of GNPs obtained by IV-OCT system.

Figure 3.2 (a) The procedure here used for the injection of GNPs into the IV-OCT system. (b) The

arrangement for “static measurements”, in which OCT images are obtained with both the GNPs solution

and catheter under static conditions. (c) The procedure followed for the acquisition of three dimensional

OCT images; The OCT catheter is moved while keeping the GNPs solution remains under static conditions.

(d) The procedure for the acquisition of OCT images of flowing GNPs; the OCT catheter is kept at fixed


58

position while the solution containing GNPs is pushed forward. (e) A representative IV-OCT slice obtained

from S-OCT video.

Static OCT imaging experiments (S-OCT): In order to evaluate the contrast enhancement

produced by the GNPs, static images where obtained, in which the IV-OCT catheter remains in

the same axial position while performing cross sectional imaging of the tubing in absence of any

flow , i.e. under static condition (see Figure 3.2 (b)). The detection limit of the IV-OCT system

for each type of GNPs was systematically investigated by injecting solutions with different GNPs

concentrations (ranging from 5×10-2 down to 5×10-6 mg/mL).

Figure 3.3 (a) S-OCT images of the conduit when it was filled with a saline solution (Control) and with

solutions containing different GNPs. In all the cases the GNP concentration was set to 0.05 mg/mL of GNPs.

(b) Integrated OCT signal as obtained for the different GNPs. (c) OCT signal normalized per the density of

GNPs as obtained for the GNPs investigated in the thesis. Note that OCT images are taken by a clinical

OCT system and so including some measuring tools (lines, reference points) that are used by clinicians that

cannot be removed from the images


59

After injection of different GNPs solutions, the static OCT videos were recorded. The static

OCT (S-OCT) cross-sectional images, which were obtained from the S-OCT videos, are shown in

Figure 3.3 (a). For the sake of comparison, the S-OCT image obtained when the tubing was filled

with only saline solution (control) is also included. From the S-OCT images included in Figure

3.3 (a), it is clear that, for all cases, the presence of GNPs in the tubing leads to a relevant

enhancement in the OCT contrast, but the amount of induced enhancement is found to be strongly

dependent on the particular type of GNP used. At this point, it should be mentioned that to estimate

the OCT enhancement, OCT intensities were first calculated, for each case, by defining a

measuring area inside the tubing in which the OCT signal was uniform. Then, using the ImageJ

software, I integrated the signal inside this area. The same procedure (same area and same

integration procedure) was applied for all the images. Thus, from these analyses, the OCT contrast

enhancement (as measured in dB, dB = 10 × log( 𝐼𝑂𝐶𝑇 𝐼𝑏𝑎𝑐𝑘𝑔𝑟𝑜𝑢𝑛𝑑)⁄ ) for each particular GNP can

be obtained. Figure 3.3 (b) shows the OCT signal enhancement as obtained from the S-OCT

images included in Figure 3.3 (a). It is clear that for the same mass concentration in the GNPs/

solution (0.05 mg/mL) GNSs present the largest OCT contrast. This is further evidenced in Figure

3.3 (c) that shows the OCT signal enhancement normalized to the GNPs concentration for each

case; the data are included in Table 3.1. Figure 3.4 shows the OCT enhancement signal per GNP

as a function of the back-scattering cross section at the OCT wavelength (given by each particular

GNP). As expected, the OCT signal enhancement monotonously increases with the back-scattering

cross section given by each type of GNPs. A linear fir to the data given in Figure 3.4, lead to an

almost linear dependence: d𝐵~𝜎𝑏𝑠𝑐𝑡0.8 . Thus, from Figure 3.4 and the data given in Table 3.1, it is

clear that, generally speaking, GNSs present scattering cross sections (at OCT wavelength) two

orders of magnitude larger than GNRs. Such difference accounts for the different OCT signal

enhancement generated by GNSs and GNRs. In particular the dominant role of back-scattering

processes in the enhancement of OCT contrast induced by GNPs is clearly manifested.


60

10-14

10-13

10-12

10-11

10-10

10-11

10-10

10-9

10-8 GNSs 986

GNSs 856

GNRs 808

GNRs 1200

GNRs 1300

GNRs 1064

back

(cm2)

dB

/GN

P

Figure 3.4 OCT signal enhancement per GNP as a function of backscattering cross section. The solid line

indicates the best linear fit to the data.

Once GNSs have been identified as the optimum GNPs ones for enhanced OCT contrast, I

proceed to evaluate the potential ability of IV-OCT detection and imaging of individual GNSs

under both static and dynamic conditions. As a first step, and in order to determine the detection

limit of our clinical OCT apparatus, I conducted systematic OCT imaging experiments using saline

solutions with different concentrations of GNSs 986 (the best scatterers). The results are

summarized in Figure 3.5. Figure 3.5 (a) shows the S-OCT images obtained for three

representative GNSs concentrations. On one hand, for elevate GNSs concentrations (0.05 mg/mL,

leading to a GNSs density of 1.2×109 GNSs/cm3) the S-OCT image is characterized by a

homogeneous and high contrast signal all along the cross section of the tubing. As one GNSs

occupying a volume of 833.33 µm3 (1×1012 µm3/1.2×109 GNSs), in this case, the movement of

single GNSs were thought to be in this volume, the shape of this volume is thought to be a cubic,

thus the size of the cubic is 833.33 µm3, the length of its side is 9.4 µm. The average separation

between GNSs was estimated to be around 9.4 µm for this concentration, which is below the axial

resolution of our OCT system (15 µm, according to the manufacturer). Therefore, for such large

concentrations, the detection of isolated/individual GNS is not possible, and the image appears as

a pseudo-homogeneous high-contrast medium. On the other hand, when the GNSs concentration

is reduced by several orders of magnitude, the OCT image appears as constituted by different

bright spikes corresponding to the local OCT enhancement caused by individual GNSs (see the

zoom of the OCT image obtained from the solution containing GNSs 986 at a concentration of

5×10-5 mg/mL). Such low concentration corresponds to a GNS density of 1.2×106 GNSs/cm3 that


61

leads to an average distance between GNSs close to 94 µm (One GNSs will occupy 8.33×105 µm3,

and has a length of 94 µm). Indeed, this interparticle distance is more than six times the nominal

axial resolution of our OCT system. Thus, under these conditions, the visualization of individual

GNSs becomes possible as it is experimentally demonstrated in the OCT images of well diluted

solutions included in Figure 3.5 (a). These results reveal the ability of nanosized individual

particles as local contrast agents for OCT and supports conclusions of previous works that pointed

out the potential capacity of OCT for imaging at the nanoscale.152-154

Figure 3.5. (a) S-OCT images of the conduit when it was filled with saline solutions containing GNSs at

different concentration levels. A detailed zoom of the S-OCT image of the 5×10-5 mg/mL concentrated

solution is also included. (b) OCT integrated intensity as a function of the concentration of GNSs. Dots are

experimental data obtained from S-OCT images and dashed line is a guide for the eyes. The background

OCT signal is indicated by the gray area. Detection limit is estimated to be 3×10-6 mg/mL. Note that OCT

images are taken by a clinical OCT system and so including some measuring tools (lines, reference points)

that are used by clinicians that cannot be removed from the images.

The dependence of the OCT intensity as a function of GNS concentration is displayed in Figure

3.5 (b), which constitutes a basis to estimate the detection limit of our OCT system for GNP

detection. The background level in our experimental conditions is also indicated by the grey area.


62

From Figure 3.5 (b) it is clear that our clinical OCT system is able to detect the presence of

colloidal GNSs with a detection limit close to 5×10-6 mg/mL that corresponds to a GNSs density

close to 1.2×105 GNS/cm3. This constitutes a relevant improvement in respect to the detection

limits previously reported for OCT imaging of GNRs and GNSs that were established to be close

to 1×1011 GNRs/cm3 and 6.2×109 GNSs/cm3.143, 151 In other words, it is here demonstrated the

superior performance of clinical OCT scans for intravascular detection of single GNPs. At this

point, such improvement is due to the special sensitivity of our experimental OCT system together

with the use of highly scattering GNSs.

3.4. Three dimensional imaging of GNPs

Three dimensional imaging of GNPs (3D-OCT): The potential application of the clinical OCT

for real three dimensional localization of GNPs was demonstrated by using the pull-back imaging

option of the clinical OCT system. In this case the fluid inside the tubing was set to be under static

conditions (no pressure applied at the end of the tubing). Then a sequence of OCT images were

recorded during pull-back of the OCT catheter over a 5 cm path length, i.e. cross-sectional images

of the tubing were measured at different axial positions within the tubing as it is schematically

indicated in Figure 3.2 (c). Subsequent analysis of the image sequence was performed by ImageJ

software, allowing for a three dimensional reconstruction of the GNP distribution along the tubing.

The obtained images were then used to estimate the density of GNPs (number of particles per unit

volume)

In other to unequivocally assign the bright spots appearing in the OCT images of low

concentrated solutions to the presence of individual GNSs, a 3D-OCT experiment was performed

along a tubing partially filled with a water solution containing, nominally, 3×107 GNSs/mL. As

schematically indicated in Figure 3.2 (c), for these particular measurements the OCT catheter was

initially positioned inside the GNSs injected volume and it was then scanned along the tubing for

a total distance of 5 mm, so that at the end of the scan the catheter reaches a tubing volume without

GNSs (i.e. just filled with saline solution). The 3D-OCT image of the GNSs was obtained by

importing the x, y and z coordinates of GNSs in the IV-OCT videos into Origin software and

drawing 3D scatter. Thus 3D-OCT image is shown in Figure 3.6 (a). Note that the interface

between the volume containing the GNSs and the volume filled with just the solution is found to


63

be close to Z = 1 mm. From the analysis of this 3D-OCT image I have computed a total number of

GNSs close to 4.2×104. The volume containing those GNPs is estimated to be 8.6×10-3 mL, leading

to a GNSs concentration of 0.5×107 GNSs/cm3. This is of the same order of magnitude, but six

times lower, than the nominal concentration of the injected solution (3×107 GNSs/cm3). The

relative good agreement between measured and nominal GNP concentration results reasonable,

specially taking into account that when measuring the 3D-OCT images the catheter is moving

along the axial (Z) direction while scanning radially the xy plane so that it is doing a spiral-shape

3D scan that would result in a slight underestimation of the actual GNSs concentration. Figure 3.6

(b) shows a detailed zoom of the 3D-OCT image in the surroundings of the water-GNSs solution

interface. From this figure, it is evidenced the nonhomogeneous distribution of GNSs along this

interface. From Figure 3.6 (b) (in which the z scale has been enlarged) it is also possible to localize

well-isolated GNSs behind the interface, i.e. in the pure aqueous volume (see red arrows). Data

included in Figure 3.6 demonstrate the potentiality of clinical OCT cardiovascular systems for the

3D localization of single GNPs.

Figure 3.6 (a) 3D-OCT image of the conduit partially filled with an injection of a saline solution containing

GNSs 856 at a concentration of 5×10-4 mg/mL. Each pixel providing a OCT signal larger than background

level has been identified as a single GNS, indicated by a dot. (b) Detailed 3D-OCT image of the interface

between the GNS solution and water. The presence of individual and isolated GNSs in the non-injected

volume of the tubing are evidenced by red arrows.


64

3.5. OCT imaging of flowing GNPs

OCT imaging of flowing GNPs (F-OCT): In order to demonstrate the potential ability of the IV-

OCT system under dynamical conditions, a pressure syringe filled with the isotonic saline solution

was connected to the end of the tubing, containing both the OCT catheter and the injected solution

of GNPs. This syringe allowed us to apply, in a controlled way, a given pressure in one end of the

tubing and, hence, an intra-tubing flow, as it is schematically indicated in Figure 3.2 (d). As a

consequence of this flow, the solution containing the GNPs was pushed from one extreme to the

other extreme of the tubing. In these experiments, the OCT catheter remained in a fixed position.

A 5 s video was recorded during the flushing out procedure of the GNPs. Analysis of this video

by ImageJ software allowed us to plot the time evolution of the GNPs distribution at the position

of the OCT catheter.

Figure 3.7 (a) F-OCT images of the conduit when it was filled with saline solution containing GNSs 986

at a concentration of 0.05 mg/mL as obtained at different times. (b) OCT integrated intensity as a function

of time during the F-OCT procedure. Dots are experimental data obtained from S-OCT images and dashed

line is a guide for the eyes. (c) Time evolution of the cross sectional OCT image during the F-OCT

procedure as obtained with a GNSs 986 concentration of 5×10-6 mg/mL.

Figure 3.7 (a) shows three representative F-OCT images corresponding to a 0.05 mg/mL

concentrated water solution of GNSs 986, as obtained at different times after establishment of an

intra-tubing evacuating flow. In this figure t = 0 s, the interface between the water and GNSs water


65

solution start crossing the OCT cross sectional detection plane. Before t = 0 s, the F-OCT image

showed a homogeneous bright signal, indicating the presence of homogeneously distributed

flowing GNSs inside the tubing. Just after t = 0 (t = 0.1 s) the F-OCT signal becomes highly

inhomogeneous, indicating that, at the position of OCT catheter, the GNSs are no longer

homogeneously distributed. Indeed, this reveals that the solution containing the GNRs at the

interface is mixing with water in a non-homogeneous way, as it was already evidenced in Figure

3.6 (c). For times larger than 0.2 s, the F-OCT contrast signal has been completely removed, as

can be observed in Figure 3.7 (b). The complete suppression of OCT contrast signal unequivocally

indicates that the solution containing the GNSs has been fully replaced by pure water and, thus, all

GNSs have been pushed out from the cross sectional OCT detection section. Appropriate analysis

of the F-OCT images obtained under flowing conditions allowed us to plot the position of single

GNSs in a certain cross-sectional area as function of time. A representative of plot is included in

Figure 3.7 (c). It can be observed how GNSs have been removed after 0.2 s pushing from the

studied cross-sectional image. Thus constitutes a solid proof of the potential ability of intravascular

OCT systems for detecting single GNPs under non-static conditions. This ability could be very

useful for a great variety of applications, including the localization and evaluation of intravascular

obstructions via detection of flowing GNPs, the accurate measurement of blood velocity and, even,

the indirect determination of intravascular pressure gradients.

3.6. Conclusions

In summary, I have experimentally demonstrated how gold nanoshells showing multipolar

plasmonic excitations in the 1000-1500 nm spectral range are excellent contrast agents for

frequency based intravascular OCT catheters, working at 1320 nm. This ability is mostly based in

the large scattering (backscattering) cross section values that these nanoparticles provide at this

specific wavelength, which are two orders of magnitude larger than those given by the gold nano-

rods traditionally used as OCT enhancers. The outstanding OCT contrast provided by gold

nanoshells has allowed for detecting individual nanoparticles under both static and dynamical

conditions. I firmly believe that the results here presented open a new avenue for novel potential

applications of intravascular optical coherence tomography.


66

On the other hand, it is important to stand out the high sensitivity of IV-OCT system. It can be

used for visualizing individual NPs both in static and dynamical conditions. This has open an

avenue of possibilities for future work.

Chapter 4. Experimental

evaluated scattering

properties of gold

nanoparticles

Chapter 4. Experimental evaluated scattering properties of gold nanoparticles

68

4.1. Introduction

After the selection of GNPs, the first step for the functional validation of these GNPs as contrast

agents for IV-OCT is a full understanding of their infrared optical properties including the

wavelength dependence of their absorption and scattering efficiencies. Such knowledge is

imperative for of IV-OCT images analysis, as well as to evaluate the thermal loading during

intravascular image acquisition. Until now there are only few experimental studies dealing with

the scattering/absorption efficiency of GNPs in the BWs. Remind that the efficiency that were

previously determined in Chapter 3 was based on the spectral simulation. Indeed, most of the

studies describing optical properties of GNPs are dealing with their response in the visible domain

or in the I-BW.155-157 The geometry dependence of the GNPs scattering efficiency has been

experimentally determined at 808 nm,158 as this one is the working wavelength OCT systems for

ophthalmology.159, 160 There are numerous predictive models on the optical response of different

GNPs in the II-BW. However, an experimental validation of the wavelength dependence of both

scattering and absorption efficiencies of GNPs in the BWs has not been demonstrated yet.

Providing this experimental data is crucial for the functional validation of GNPs as contrast agents

for advanced IV-OCT.

In this chapter, the scattering efficiency of three types of GNPs (gold nanorods, gold nanoshells

and gold nanostars) displaying LSPR with significant extinction at around 1.3 µm (the operating

wavelength of clinical IV-OCT apparatus) have been experimentally determined. By means of a

well-stablished photo-thermal method,126, 157, 161 the scattering efficiency has been measured at

three different wavelength, namely at 808 nm, 980 nm and 1280 nm. These multi-wavelength

experimental data have been compared with numerical simulations performed in Chapter 3, we

will see that simulation failed to reproduce reliable values at long wavelengths. This fact stands

out the importance of an experimental determination of the scattering and absorption efficiencies.

The geometry dependence of scattering efficiency determined by the photo-thermal method has

been corroborated by the combination of IV-OCT, infrared dark field microscopy (IR-DFM) and

infrared optoacoustic (IR-OA) spectroscopy. The optimum GNPs geometry for IV-OCT contrast

enhancement has been determined and the importance of the experimental determination of the

spectral dependence of infrared scattering efficiency has been clearly evidenced.


69

4.2. Characterization of the GNPs

Figure 4.1 (a) TEM images of the GNSs, GNSTs and GNRs used along this chapter. (b) Size distribution

of GNSs, GNSTs and GNRs

Three kind of the GNPs with different shape were used in this chapter; GNRs, GNSs and gold

nanostars (GNSTs). GNRs and GNSs were the same describe before, whereas GNSTs were

synthesized at Pompeu Fabra University (UPF). Figure 4.1 shows the representative TEM images

and size distribution of these GNPs. Here, the TEM of the GNRs and GNSs are shown again due

to the new batches of GNRs and GNSs were used here (their TEM images were previously shown

in Chapter 3). GNRs have a diameter of 11±2 nm and a length of 90±20 nm. GNSs have a total

average diameter of 220±23 nm. GNSTs consist in solid gold nanospheres with an average

diameter of 28±4 nm together with different lobes (in average 14) with an average length of 14±3

nm.


70

The extinction cross section spectra of these three structures are included in Figure 4.2. The

extinction cross section spectra shape of GNRs and GNSs are the same as they were described in

Chapter 3. The extinction cross section of GNSTs is constituted by a plasmonic band peaking at

about 800 nm and extends in the IR spectral region from about 500 nm to 1600 nm. Such LSPRs

are electric dipole resonances stemming from the collective oscillations of Au electrons. As such,

they are sensitive to nanoparticle shape (except for very small nanoparticle), especially along the

direction of electric field polarization of the incident light. GNSTs are not spherically symmetric;

however, for a sufficiently large number of star tips pointing to all directions, they are nearly

polarization independent.162,121, 163 Due to broad extinction spectra of these GNPs, the relevant

infrared wavelengths were investigated in this chapter; 808 nm, 980 nm and 1280 nm (the IV-OCT

working wavelength).

Figure 4.2 Extinction spectra obtained from colloidal suspensions of GNSs, GNSTs and GNRs.

4.3. Experimental determination of infrared scattering efficiency

The experimental determination of the wavelength dependence of the scattering efficiency (𝜂𝑠)

is based on the experimental determination of the photo-thermal heating conversion efficiency (𝜂ℎ)

at each particular wavelength (808, 980, 1280 nm). 𝜂ℎ is defined as the fraction of extinct

(scattered or absorbed) energy by a single nanoparticle that is converted into heat. For non-

luminescent nanoparticles, as it is our case, all the absorbed energy by a single nanoparticle is

converted into heat, so that the photo-thermal conversion efficiency is equal to the absorption

efficiency (𝜂𝑎) (i.e. 𝜂ℎ = 𝜂𝑎 ), as defined in Chapter 3. At a given wavelength (𝜆), it can be

written:


71

𝜂ℎ(𝜆) = 𝜂𝑎(𝜆) =𝜎𝑎𝑏𝑠(𝜆)

𝜎𝑎𝑏𝑠(𝜆) + 𝜎𝑠𝑐𝑡(𝜆)=

𝜎𝑎𝑏𝑠(𝜆)

𝜎𝑒𝑥𝑡(𝜆), (4.1)

where 𝜎𝑎𝑏𝑠(𝜆), 𝜎𝑠𝑐𝑡(𝜆) and 𝜎𝑒𝑥𝑡(𝜆) are the absorption, scattering and extinction cross sections at

a wavelength 𝜆, respectively. If the photo-thermal conversion efficiency 𝜂ℎ is known, then the

scattering efficiency, as described in Chapter 3, can be written:

𝜂𝑠(𝜆) = 1 − 𝜂𝑎(𝜆) = 1 − 𝜂ℎ(𝜆). (4.2)

The photo-thermal conversion efficiency can be evaluated by measuring the input/output heat

balance under transient illumination, as previously reported.126, 157, 164 Figure 2.3 (a) shows the

experimental arrangement used all along this work to measure 𝜂ℎ(𝜆). The method is described in

detail in Chapter 2 Experimental techniques. It is based on the measurement of the heating and

cooling curve of a solution containing the GNPs when it is optically excited by a fiber coupled

laser diode. A representative heating/cooling curve is shown in Figure 2.3 (b). From this curve, it

is possible to estimate the absorption efficiency, which it is given by:

𝜂𝑎 = 1 − 𝜂𝑠 =

𝑚𝐶𝑝∆𝑇𝑚𝑎𝑥

𝜏 − 𝑄𝑟

𝑃0(1 − 10−𝑂𝐷), (4.3)

where 𝑚, 𝐶𝑝 and 𝑂𝐷 are the mass, specific heat and optical density of the solution containing the

GNPs. 𝑃0 is the incident laser power, ∆𝑇𝑚𝑎𝑥 is the maximum temperature increment (indicated in

Figure 2.3 (b)) and 𝜏 is the relaxation time of the cooling process (also indicated in Figure 2.3

(b)). Equation (4.3) gives the photo-thermal efficiency (equal to the absorption efficiency) from

parameters that are either known or can be experimentally measured. For instance, the specific

data shown Figure 2.3 (b) corresponds to the heating and cooling cycle of an aqueous solution of

GNRs excited at 1280 nm and for the particular values m = 0.5 g, P0 = 160 mW, OD = 1.075,

ΔTmax = 9.92 ºC and τ = 104.3 s. Substituting these values in Equation (4.3), an absorption

efficiency of 0.95 was obtained, and therefore, a scattering efficiency of ƞs: 1-0.95 = 0.05.


72

Figure 4.3 Cooling curves of the GNPs under excitation at different wavelength.

Table 4.1 Parameters to calculate absorption efficiency

Laser

wavelengt

h (nm)

P0

(mW)

Qr

(mW) s

GNSs

OD

GNSs

ΔT(°C)

GNSs η(±10%)

GNSs

Average

η

Simulated η

GNSs

1280

125 45 120.5

0.406

4.81 51

49 32 142 52 126.0 5.85 52

160 60 126.8 6.31 45

980

245 70 135.12

0.67

6.55 16

16 22 280 80 129.91 7.41 18

310 100 124.83 8.12 15

808

180 16 96.00

0.55

2.97 38

32 28 218 18 118.00 3.61 29

257 22 118.50 4.21 28


73

Laser

wavelength

(nm)

P0

(mW) Qr

(mW) s

GNRs

OD GNRs

ΔT(°C) GNRs

η(±10%)

GNRs

Average

η

Simulated η

GNRs

1280

125 45 121.40

1.075

8.07 82

88 95 142 52 114.00 9.11 88

160 60 104.30 9.92 95

980

245 70 123.60

0.25

8.53 69

77 95 280 80 116.72 10.43 87

310 100 121.73 11.65 74

808

180 16 130.00

0.125

3.60 93

95 95 218 18 124.10 4.61 100

257 22 124.80 4.78 91

Laser

wavelength

(nm)

P0

(mW)

Qr

(mW) s

GNSTs

OD

GNSTs

ΔT(°C)

GNSTs η(±10%)

GNSTs

Average

η Simulated η

GNSTs

1280

115 35 90.97

1.06

4.5 65

56 91 130 45 101.28 5.07 49

148 55 89.42 5.52 55

980

155 32 77.17

3.9

6.28 89

84 87 350 80 73.82 13.06 83

540 120 73.87 19.53 80

808

187 18 79.08

4.835

7.62 98

90 87 290 40 79.43 11.4 90

420 50 80.14 15.48 84

In order to obtain an accurate statistical value of the scattering efficiency, experiments were

carried out at different input powers for each one of the three illumination wavelengths here

investigated (see Figure 4.3 and Table 4.1). Note that once the scattering efficiency at a given

wavelength (𝜆) is experimentally determined, it is possible to get the scattering extinction cross

section at this wavelength, since:

𝜎𝑠𝑐𝑡(𝜆) = 𝜂𝑠𝑐𝑡(𝜆) · 𝜎𝑒𝑥𝑡(𝜆), (4.4)

where 𝜎𝑒𝑥𝑡(𝜆) is the extinction cross section at wavelength 𝜆 as experimentally determined from

the extinction spectra included in Figure 4.2. In Table 4.2, the averaged scattering efficiencies,


74

extinction cross section and scattering cross section as obtained for each GNP (at the three

wavelengths) are listed. Inspection of Table 4.2 reveals that the largest scattering efficiencies and

cross sections are provided by GNSs. Actually, the scattering efficiency of GNSs is higher than

50%, for the three wavelengths investigated in this work. Consequently, these results indicate that

GNSs are the most suitable nanoparticles for scattering based imaging techniques, such as OCT

and optical microscopy.127, 165 This is in good agreement with the scattering efficiency estimated

by means of simulation in Chapter 3. Data included in Table 4.2 also reveal that the scattering

efficiency is strongly wavelength dependent for all GNPs. Although GNSTs show very low

scattering efficiencies at 800 and 980 nm they present a scattering efficiency close to 50% at

around 1.3 µm.

Table 4.2. Experimentally determined scattering efficiency and scattering cross sections at three

wavelengths investigated.

808 nm 980 nm 1280 nm

ηsct σext (cm2) σsct (cm2) ηsct σext (cm2) σsct (cm2) ηsct σext (cm2) σsct (cm2)

GNSs 0.68 1.50×10-9 1.02×10-9 0.84 1.81×10-9 1.52×10-9 0.51 1.09×10-9 5.56×10-10

GNRs 0.05 1.64×10-12 0.82×10-13 0.23 5.00×10-12 1.15×10-12 0.12 1.42×10-11 1.70×10-12

GNSTs 0.10 2.55×10-11 2.55×10-12 0.16 2.06×10-11 3.30×10-12 0.44 5.59×10-12 2.46×10-12

4.4. Numerical calculations and comparison with experimental data

In this chapter, new numerical calculations were provided by Diego Romero and José A.

Sánchez-Gil (Instituto de Estructura de la Materia (IEM-CSIC), Consejo Superior de

Investigaciones Cientificas). The calculation process is illustrated below.

Full electrodynamic methods were used to calculate the spectral dependence of the extinction,

absorption and scattering cross sections. For GNSs, the simulation method is the same as described

in Chapter 3. For more complex shapes, such as GNRs and GNSTs, numerical calculations were

carried out by means of a free software implementation, called SCUFF166 (open-source software

package for analysis of electromagnetic scattering problems using the method of moments). Gold

permittivity data were obtained from Johnson and Christy measurements167 and the medium


75

permittivity is assumed to be the water (n = 1.33). Thus, GNRs were considered as cylinders with

given length and radius, with endcaps terminated as hemispheres with the same radius.

For the particular case of GNSTs the situation is more complex and a crucial issue is how to

define the nanostar geometry. In this regard, the definition of a nanostar shape is inspired by the

TEM images (Figure 4.1). Rather than exploiting analytical formulation (as in Ref.162, which poses

some difficulties in the meshing procedure), each nanostars is assumed contain a spherical core,

with fourteen elongated tips with semi-ellipsoidal shape, as shown in Figure 4.4 (a). Due to the

structural inhomogeneities shown in the TEM images of the GNTSs, the spherical core radius was

fixed as R = 15 nm, and the tip ellipsoidal cross section (along two short axes) as a circle with

radius r = 5 nm. The tip length L (long axis of the ellipsoid) is then allowed to vary between 12-

25 nm (which nearly corresponds to the tip length and angle (θ = atan(r/L)), as observed in the

TEM images. The resulting calculated cross sectional spectra are shown in Figure 4.4 (b) for

monochromatic plane waves, with varying vacuum wavelength impinging onto the nanostars along

the x direction, and with linear polarization along the vertical axis (with respect to the nanostar

shown in the inset). Strong LSPRs appear with narrow width in the range 700-950 nm, with a ratio

between absorption and extinction of ~ 87%. In this regard, it should be emphasized that, despite

the anisotropic shape of the nanostars, the resulting spectra are nearly isotropic and polarization

independent. This was verified (though not shown) by varying the angle of incidence and the

electric field polarization. Actually, this was also pointed out in reference,163 even for smoother

nanostars with fewer tips. The reason is that the LSPR, being essentially a dipolar resonance, is

excited mostly at any angle of incidence, due to fact that the electric field vector stretches along

the direction between any of the opposing nanostar tips, whose length (~2R+2L) in turn controls

the LSPR wavelength. As a matter of fact, upon the nanostar tip length increases, the LSPR is red-

shifted, covering the spectral range between 700-950 nm, consistently with the broad LSPR width

measured experimentally (Figure 4.2). To shed light on such inhomogeneous broadening, a

smoothed weighted average was carried out over all the spectra of different tip lengths, assigning

approximate weighs according to the statistics of GNSTs shapes. The resulting simulated spectra

have been included in Figure 4.4.


76

Figure 4.4 (a) Schematic draw of a nanostar, containing an spherical core with fourteen elongated tips with

semi-ellipsoidal shape. (b) Calculated extinction and absorption spectra considering a spherical core radius

R = 15 nm, with fourteen elongated tips with semi-ellipsoidal shape with r = 5 nm short axes and varying

long axis L = 12, 13.5, 15, 16.5, 18, 20, 25 nm. A weighted average over all the spectra for different tip

lengths is also shown (smoothed over ±5 nm), with approximate weighs extracted from the statistics of

GNST shapes (TEM images, Figure 4.1): L = 12 (13%), 13.5(22%), 15(22%), 16.5(22%), 18(22%),

20(17%), 25(4.4%) nm.

Figure 4.5 Experimental extinction, calculated extinction, absorption and scattering cross section spectra

of GNSs, GNRs and GNSTs in the I and II-BW.

The numerically calculated extinction, absorption and scattering cross section spectra in the

spectral range 400-1600 nm (i.e covering BWs) are included in Figure 4.5 as calculated for GNSs,

GNRs and GNSTs. By comparing these data with the experimentally obtained extinction spectra,

GNSs are in good agreement, while the spectral shapes are not properly reproduced neither for

GNRs nor for GNSTs. For GNRs the calculated extinction cross section spectrum (peaking at about


77

1300 nm) appears shifted to longer wavelengths in respect to the experimental one (peaking at

about 1200 nm). In addition, the calculated spectrum is narrower than the experimental one. For

GNSTs, while the peak position is certainly reproduced, the calculated spectrum appears much

narrower than the experimental one, and so leading to a large inaccuracy when determining the

cross section values and the scattering efficiencies at long wavelengths. This experimental

broadening is likely due to the non-uniform nanostar geometries (e.g. wide distribution of aspect

ratio), as well as the finite-size effects, which broaden and dampen the resonance but do not

influence the plasmon band position.121 It should be borne in mind that this experimental

broadening has been reported elsewhere also in connection with star-shaped nanoparticles,121 as

stemming from both the nonuniform nanostar geometries, and finite-size effects at the tip edges

(shorter electron scattering mean free path). The former mechanism has been somehow taken into

account through the average over GNSTs shapes inspired by the TEM images (Figure 4.1 (a)).

Nevertheless larger shape variations and/or particle aggregation (not evidenced in the TEM images)

might be responsible for the discrepancies. In addition, the other mechanism mentioned above,

finite-size effects, also typically broadens resonance; it is due to increased electron damping

resulting from electron scattering at the narrow tips, similar to what occurrs within GNSs thin

shells.168, 169

Figure 4.6 Wavelength dependence of scattering efficiency of GNSs, GNSTs and GNRs. Solid line shows

results extracted from calculations whereas the dots correspond to the experimental data obtained in this

work (see Table 4.2).

By using the calculations included in Figure 4.5, the wavelength dependence of scattering

efficiency can be also numerically calculated for all the investigated GNPs. Results of these


78

calculations are included in Figure 4.6 together with here obtained the experimental data at the

three used wavelength. According to these numerical calculations, the scattering efficiency is

weakly dependent on wavelength for GNRs and GNSTs but strongly wavelength dependent for

GNSs. While a good agreement between numerical simulations and measured efficiencies occurs

at 800 nm and 980 nm, a clear disagreement is observed at around 1.3 µm (the IV-OCT wavelength)

for GNSTs. At this specific wavelength, calculations predict a scattering efficiency of 0.09, much

lower than that experimentally measured (0.44). While the origin of this discrepancy is not fully

understood, the physical phenomenon behind this must be the same that leads to the appearance

of the long wavelength tail in the extinction spectrum. It was assumed that the existence of

interaction between individual GNSTs via agglomeration or slight contact between tips, could lead

to dimer structures with completely different optical properties. In particular, the presence of such

interacting structures could lead to red-shifts in the plasmon resonance and to larger scattering

efficiencies. Regardless of the origin of these discrepancies, this highlight the importance of

performing direct experimental measurements for the scattering efficiency at different

wavelengths, as it was done in this chapter.

4.5. Infrared scattering experiments

To gain insight into the actual scattering efficiency of GNPs at long wavelengths, where the

calculated and measured data display the largest discrepancy, two set of imaging measurements

were carried out. In particular I investigated the infrared scattering abilities of the three types of

GNPs by infrared dark field microscopy (IR-DFM) and IV-OCT. The results obtained in each case

are described in the following.


79

4.5.1. Infrared dark field microscopy

Figure 4.7 (a) Schematic representation of the experimental set up designed to evaluate the ability of GNSs,

GNSTs and GNRs as contrast agents in a dark field configuration. (b) Band pass spectra around 1300 nm

after substract the long pass light after 1250 nm by long pass light after 1350 nm.

For IR-DFM experiments the GNPs were deposited and dried onto the microscope slides. The

microscope slides containing the GNPs were visualized by using the home-made IR-DFM, the

schematically was drawn in Figure 4.7 (a). In order to just get the infrared DFM Image (i.e. just

record the dispersed light around 1280 nm), two long pass filters of cutting wavelengths 1250 nm

and 1350 nm were used (see the experiemtal set up in Figure 4.7 (a)). In detail, the following

transmission spectra; (1) after putting long pass filter 1250 nm and (2) after putting long pass filter

1350 nm were obtained. In such a way, a narrow band pass spectrum peaking around 1300 nm was

obtained after substract (1) by (2), as shown in Figure 4.7 (b). Therefore, only the infrared radiaton

sattered by the GNPs in the wavelength range from 1250 nm to 1350 nm was obtained.

Consequently, the here employed IR-DFM system is able to visualize individual GNPs by using

their scattering abilities at 1300±50 nm. Figure 4.8 (a) shows the IR-DFM images obtained for

the three GNPs studied in this chapter. Up to the best of my knowledge, these are the first ever

reported IR-DFM images of GNPs in the II-BW. As can be observed, all GNPs can be detected by

IR-DFM, revealing a non-vanishing scattering cross section in the three cases (in accordance with


80

the experimental data included in Figure 4.6). The largest IR-DFM contrast has been provided by

GNSs, being significantly larger than those provided by GNSTs and GNRs. This is more clearly

evidenced in Figure 4.8, in which I have included the average IR-DFM intensity per spot generated

by GNSs, GNRs and GNSTs. The results included in Figure 4.8 are, therefore, in qualitative

agreement with the experimentally determined and numerically calculated values of the scattering

cross section (see Table 4.2). These experimental data confirm that GNSs provide the largest

infrared scattering cross section among the all geometries here studied.

Figure 4.8 (a) Infrared dark field images obtained at 1.3 µm by using GNSs, GNRs and GNSTs as potential

contrast agents. Scale bar is 20 µm in all the cases. (b) Average intensity per nanoparticle generated by

GNSs, GNRs and GNSTs as obtained from the statistical analysis of IR-DFM images.

4.5.2. Intravascular optical coherence tomography

The infrared scattering ability of the different GNPs at around 1.3 µm (the IV-OCT laser

wavelength) has also been investigated by performing IV-OCT experiments. In this case, a

colloidal solution of each type of GNP was injected into a tubing (see Experimental techniques).

Once the aqueous solutions of GNPs were inside the tubing (mimicking an artery), the IV-OCT


81

catheter was introduced into the tubing allowing for the static detection of the individual GNPs,

by means of their back-scattering efficiency.24 It should be noted that the advantage of this

interferometric technique OCT is that it provides an indirect measure of the infrared back-

scattering cross section at the single wavelength of 1.3 µm. Figure 4.9 shows the IV-OCT cross

sectional images for the colloidal solutions containing the three investigated GNPs. Again, the

three geometries here studied produced a significant IV-OCT signal. This again reveals a non-

vanishing infrared scattering cross section at around 1.3 µm. It is also evident from Figure 4.9 (a)

that the IV-OCT contrast is provided by each type of GNP. Figure 4.9 (b) includes the averaged

OCT intensity (contrast) provided by individual GNPs, as obtained for each type of GNP. Again,

and as we also obtained in Chapter 3, it is found that the largest contrast is provided by GNSs,

revealing their superior scattering cross section.

Figure 4.9 (a) OCT cross section images of colloidal solutions containing GNSs, GNSTs and GNR. (b)

OCT intensities of single GNSs, GNSTs and GNRs obtained from the statistical analysis of figures included

at the left column.

At this point, the concordance between experimental data included in Figure 4.9 and those of

Table 4.2 can be discussed. For a sample of thickness d containing a certain numer of GNPs, i.e.

N, the Lambert-Beer exctinction law gives the fraction of scattered intensity:

𝐼0 − 𝐼

𝐼0= 1 − 𝑒−𝜎𝑠𝑐𝑡𝑁𝑑, (4.5)


82

where I0 is the incident light intensity and I the light intensity transmitted through the sample. For

low values of the exponential exponent, and assuming the same thickness for all measurements

(same geometrical conditions), Equation (4.5) can be written as:

𝐼0 − 𝐼

𝐼0≈ 𝜎𝑠𝑐𝑡𝑁𝑑, (4.6)

thus the OCT intensities (per GNP) given in Figure 4.9 (b) should be proportional to the fraction

of scattered light and so to experimental values of 𝜎𝑠𝑐𝑡 (indeed to the backscattering cross section)

given in Table 4.2. Now, the OCT intensity per GNP (Ioct,) (for each type of GNP) normalized to

that given by the GNSs (the best scatterers) was compared. From the data given in Figure 4.9, I

obtained:

𝐼𝑜𝑐𝑡(𝐺𝑁𝑆𝑇𝑠)

𝐼𝑜𝑐𝑡(𝐺𝑁𝑆𝑠)⁄ = 5 × 10−3 (4.7)

and

𝐼𝑜𝑐𝑡(𝐺𝑁𝑅𝑠)

𝐼𝑜𝑐𝑡(𝐺𝑁𝑆𝑠)⁄ = 2 × 10−3. (4.8)

These values are indeed close to the ratio between the scattering cross sections experimentally

obtained and listed in Table 4.2, that leads to: 𝜎𝑠𝑐𝑡(𝐺𝑁𝑆𝑇𝑠) 𝜎𝑠𝑐𝑡(𝐺𝑁𝑆𝑠)⁄ ≈ 4.42 ×

10−3and 𝜎𝑠𝑐𝑡(𝐺𝑁𝑅𝑠) 𝜎𝑠𝑐𝑡(𝐺𝑁𝑆𝑠)⁄ ≈ 3.05 × 10−3. In fact, the slight discrepancy could be due

to the different geometries that lead to different backscattering/scattering ratios. Therefore, at this

point, the IV-OCT experimental results confirm the validity of the scattering efficiencies (and so

the scattering cross sections) experimentally determined in this thesis by the photo-thermal method.

In fact, the results here reported highlight the need of a direct experimental determination of the

scattering efficiency, particularly at longer wavelengths (as for the OCT wavelength), where

numerical calculations for GNRs and GNSTs fail to explain the experimentally obtained extinction

cross section (see Figure 4.2 and Figure 4.5). Therefore, the experimental determined scattering

efficiencies are crucial to provide reliable values.

4.5.3. Optoacoustic experiments

An indirect estimation of the geometry dependence of the scattering efficiency of GNPs can be

also obtained by the performance of IR-OA experiments.170 For this purpose I prepared colloidal


83

solutions containing the three types of GNPs with exactly the same OD at 1064 nm (this being the

wavelength used in our IR-OA experiments), as shown in Figure 4.10 (a). Each solution was

introduced in a quartz cuvette mechanically attached to a piezo transducer, connected to signal

amplifier and to a digital oscilloscope. In absence of any fluorescence (as in our case) the

magnitude of the IR-OA signal is proportional to the fraction of absorbed pump power (𝑃𝑎𝑏𝑠 =

𝑃0 · (1 − 10−𝑂𝐷)), where 𝑃0 is the incident pump power that is converted into heat. In other words,

the OA signal is expected to be proportional to 𝜂𝑎𝑏𝑠 · 𝑃𝑎𝑏𝑠. Therefore, we can write:

𝐼𝑂𝐴 = 𝑘 · (1 − 𝜂𝑠𝑐𝑡) · 𝑃0 · (1 − 10−𝑂𝐷), (4.9)

where 𝑘 is a coupling constant that depends on the Grüneisen parameter and the experimental

conditions such as the gain of the piezo transducer.171 From Equation (4.9), it is clear that the IR-

OA signal is proportional to (1 − 𝜂𝑠𝑐𝑡) in such a way that those GNPs with largest scattering

efficiencies and with the same OD should give the lowest IR-OA signal. Figure 4.10 (b) shows

the IR-OA signals generated by the three GNPs studied in this work. The three geometries generate

measurable IR-OA signals even though with quite different amplitudes. The averaged IR-OA

amplitudes obtained for the three cases are displayed in Figure 4.10 (c). In this case, the lowest

IR-OA signal was generated by the GNSs. This is in close agreement with the experimental data

displayed in Figure 4.6 and with Equation (4.9) that predicts the lowest IR-OA for the GNPs with

the largest scattering efficiency. Data included in Figure 4.10 (c) allows for a quantitative

estimation of the relation between the infrared scattering efficiencies of the different GNPs studied

in this work. Equation (4.9) can be rewritten to get a formal expression for the OA intensity ratio

between GNSTs and GNSs, so that:

𝑅𝑂𝐴 (𝐺𝑁𝑆𝑇𝑠

𝐺𝑁𝑆𝑠) =

𝐼𝑂𝐴(𝐺𝑁𝑆𝑇𝑠)

𝐼𝑂𝐴(𝐺𝑁𝑆𝑠)=

1 − 𝜂𝑠𝑐𝑡(𝐺𝑁𝑆𝑇𝑠)

1 − 𝜂𝑠𝑐𝑡(𝐺𝑁𝑆𝑠). (4.10)

Analogously, Equation (4.10) can be written for the ratio that compares GNRs and GNSs, as

follows:

𝑅𝑂𝐴 (𝐺𝑁𝑅𝑠

𝐺𝑁𝑆𝑠) =

𝐼𝑂𝐴(𝐺𝑁𝑅𝑠)

𝐼𝑂𝐴(𝐺𝑁𝑆𝑠)=

1 − 𝜂𝑠𝑐𝑡(𝐺𝑁𝑅𝑠)

1 − 𝜂𝑠𝑐𝑡(𝐺𝑁𝑆𝑠). (4.11)


84

From the experimental IR-OA data, I obtained 𝑅𝑂𝐴 (𝐺𝑁𝑆𝑇𝑠

𝐺𝑁𝑆𝑠) ≈ 𝑅𝑂𝐴 (

𝐺𝑁𝑅𝑠

𝐺𝑁𝑆𝑠) ≈ 6 . From the

experimentally determined scattering efficiencies (Table 4.2), I obtained 𝑅𝑂𝐴 (𝐺𝑁𝑆𝑇𝑠

𝐺𝑁𝑆𝑠) ≈

𝑅𝑂𝐴 (𝐺𝑁𝑅𝑠

𝐺𝑁𝑆𝑠) ≈ 4. Therefore, it is found that the infrared scattering efficiencies determined by the

transient photo-thermal method, are in good agreement with optoacoustic experiments of the GNPs

studied in this work.

Figure 4.10 (a) Optical density of GNRs (red), GNSTs (green) and GNSs (blue). (b) Time evolution of the

IR-OA signals generated by the three GNPs after laser excitation. (c) Average IR-OA amplitudes obtained

for the three GNPs.

4.6. Conclusions

An experimentally determination of infrared scattering efficiency and scattering cross section

of GNPs with three different geometries: nanorods, nanoshells and nanostars was carried out in

this chapter. Multi-wavelength thermal loading experiments have been used to experimentally

determine the spectral dependence of the scattering efficiency of these three geometries.

Comparison between experimental data and numerical simulation, reveals good agreement in all

cases, except for GNSTs that present an unexpected large scattering efficiency at long wavelengths.

This discrepancy has been explained in terms of size inhomogeneities as well as particle-particle

interactions. Both GNSs and GNSTs show scattering efficiencies strongly dependent on the

wavelength, thus evidencing the importance of considering the spectral operating range when

choosing the appropriate geometry for each particular imaging application. It is also

experimentally found that, in the whole infrared spectral range investigated in this work (800-1400


85

nm), GNSs show the largest scattering efficiency and cross section. This explains, why they are

best cardiovascular imaging probes, as we also concluded in Chapter 3.

The superior performance of GNSs as scattering units in the infrared has been confirmed by

means of infrared dark field imaging, optical coherence tomography and optoacoustic experiments.

In all the cases, the experimental data confirm the previous conclusions extracted from photo-

thermal experiments. They also reveal the multifunctional character of these GNPs that are capable

of efficient light scattering and acoustic generation under infrared optical excitation.

Chapter 5. GNSs as contrast

agent for cell imaging and

tissue contrast enhancement

by OCT

Chapter 5. GNSs as contrast agent for cell imaging and tissue contrast enhancement by OCT

88

5.1. Introduction

IV-OCT imaging is nowadays used as a minimally invasive technique for the in vivo

identification and assessment of unstable atheroma plaques as well as for the guidance of

interventional procedures, such as atherectomy and stent placement.24, 172-175 The possibility of

adding molecular contrast to available IV-OCT systems opens a new door to extend the impact of

OCT to cardiovascular clinical applications, as it would enable functional biomolecular imaging

together with available morphological imaging. Indeed, IV-OCT with molecular contrast would

enable site-specific cardiovascular imaging and therefore would allow for the early detection and

localization of atherosclerotic lesions, which is not possible by using current IV-OCT imaging

techniques. Thus, this intravascular molecular imaging modality would contribute substantially to

the future development of personalized medicine for cardiovascular diseases.

A common strategy to improve OCT contrast at the molecular level involves the use of

nanoparticles (NPs) that enhance OCT contrast and, at the same time, can be properly

functionalized with antibodies to target specific molecules overexpressed by the tissue/cells to be

detected.106, 176 In this respect, GNPs of different geometries and sizes have been successfully

applied as OCT contrast enhancers.82, 149, 177-181 GNPs are particularly suitable for OCT contrast

improvement as their extinction spectra can be precisely tuned over a broad spectral range by

solely controlling their size and geometry. In addition, the gold surface of GNPs is particularly

suitable for functionalization with specific targeting ligands and therefore provides the opportunity

for molecular targeting.181 Owing to these properties, GNPs have already been deployed as contrast

agents in different OCT imaging modalities.81, 182, 183 Moreover, as gold is an inert and

biocompatible noble metal, these GNPs are considered nontoxic, although some concerns have

been raised, especially related to the long-term effect of gold accumulation in the organism or

cellular modifications.184

In Chapter 3, it was described how different types of commercial GNPs, including gold

nanorods and gold nanoshells (GNSs), with plasmon resonances in the proximity of 1320 nm,

could act as efficient IV-OCT enhancers when suspended in biomedical fluids.179 In particular,

GNSs consisting of a dielectric silica core of about 200 nm and an ultrathin metal shell of about

15 nm, resulted in the largest IV-OCT contrast enhancement based on their superior backscattering


89

cross-section (0.7 × 10-10 cm2) at the IV-OCT wavelength.179 As a result, GNSs dispersed in

physiological solutions could be individually observed by IV-OCT, even under dynamic flow

conditions. Despite these promising features, the potential use of GNSs for the selective

enhancement of live cell imaging by IV-OCT must be explored, though it would constitute the

first step towards the use of IV-OCT for molecular imaging.

In this chapter, a systematic investigation of IV-OCT contrast in living cells and its enhancement

by passive incubation with GNSs is presented. Experiments were conducted in two different cell

lines; HeLa and Jurkat cells. (i) HeLa cells constitute an immortal cell line, as they can divide

many times in cell culture plate under minimal survival conditions. Thus, they have been widely

used for scientific research. What’s more, as cancer cells, they are capable to uptake NPs by

endocytosis. (ii) Jurkat cells are T lymphocytes and can be found in the blood stream. Recent

studies have shown that they play an important role in protecting against cardiovascular diseases

by responding to inflammatory cues.185, 186 Therefore, the investigation of these two cell lines for

OCT molecular imaging is important. Intensity analysis of IV-OCT was used to discriminate the

population of living cells in which an efficient incorporation of GNSs was achieved and to

determine the relationship between the incorporation efficiency and the incubation parameters.

The results extracted from the IV-OCT images of cell dispersions were compared to the images

obtained by transmission electron microscopy (TEM) and dark field microscopy (DFM), in order

to unequivocally correlate the contrast enhancement observed in the IV-OCT images with the

incorporation of GNSs in living cells. Future implications of the results for the application of IV-

OCT to intracoronary molecular imaging are also discussed.

1.1. Static OCT imaging of HeLa and Jurkat cells incubated with GNSs

As indicated in the Introduction, and with the purpose of studying IV-OCT contrast for living

cells and its enhancement by passive incubation with GNSs, cytotoxicity was initially examined

using various concentrations of GNSs incubated on HeLa cells for 24 h. MTT assay was then used

to evaluate the cytotoxicity of the GNSs, more details can be found in Experimental techniques.

The results of these experiments are shown in Figure 5.1. For all the concentrations (0.05, 0.1, 0.5,

2.5, and 5 µg/mL), the cell viability was never less than 94% relative to the control. Therefore, the

GNSs did not show any relevant toxicity to the HeLa cells under these experimental conditions.


90

Figure 5.1 Cell viability of HeLa cancer cells after incubation with several concentrations of GNSs. Each

bar corresponds to the mean value ±standard deviation (SD) obtained from 4 replicates for each

experimental condition.

Figure 5.2 (a) IV-OCT cross-sectional images of GNSs suspended in PBS , nonincubated HeLa and Jurkat

cells in PBS, HeLa and Jurkat cells incubated for 24 h with a dispersion of GNSs in PBS. (b) Histograms

of intensity, as calculated from the IV-OCT images. Note that free GNSs and free (nonincubated) Jurkat

and HeLa cells yielded single Gaussian distributions centered at about 65 uarb (GNSs) and 60 uarb (free

cells), whereas the incubation of HeLa and Jurkat cells with GNSs led to the appearance of high-intensity


91

bands centered at about 120 uarb. The different fits to Gaussian distributions are shown as solid lines. ×3

and ×10 indicate the magnification factor. (c) Representative IV-OCT cross-sectional images of HeLa cells

incubated with 5 µg/mL GNSs for 24 hours, obtained by injecting the sample (1, 2 and 3) and its duplicate

(4, 5 and 6) in the tubing at different times, and moving the IV-OCT catheter in 3 positions along the tubing.

After the negligible toxicity of the GNSs was proved, the GNSs were incubated in cells (HeLa

and Jurkat) to evaluate their viability as OCT contrast agents. HeLa cells were incubated with 2.5

µg/mL of GNSs for 24 hours, while Jurkat cells were incubated with 1.0 µg/mL of GNSs for 24

hours. Both of them were washed for three times in order to remove the excessed GNSs, and then

they were dispersed in PBS. Then 200 µL of the incubated cells/PBS solution were injected in a

tubing containing the IV-OCT catheter. In order to make a comparison, the following control

samples were also prepared and then injected into the tubing: (i) PBS solution, (ii) 2.5 µg/mL of

GNSs in PBS, (iii) non-incubated HeLa and Jurkat cells in PBS solution. Figure 5.2 (a) shows

OCT cross-sectional images of the tubing filled with these solutions. As we have seen in Chapter

3, individual GNSs can be visualized by IV-OCT when using a low concentration of GNSs. The

different bright spots in the cross-sectional images are due to the backscattered light of the OCT

laser produced by (i) individual GNSs, (ii) individual cell, (iii) cell containing GNSs. Thus, the

cross-sectional images in Figure 5.2 (a) must be carefully analyzed. For this purpose, the spot

intensity distribution of these cross-sectional images was carefully analyzed. Figure 5.2 (b) shows

the spot intensity distribution (spot frequency as a function of the OCT spot intensity) obtained for

the different samples. The data analysis process is described in detail in Experimental techniques.

Each distribution was obtained by analyzing the intensity of 120 images taken along the tubing.

To guarantee the accuracy, the experiment was repeated at least 6 times for each concentration by

injecting each sample and its duplicate (another solution with same GNSs concentration) in the

tubing, and moving the IV-OCT catheter to 3 different positions along the tubing. An example of

these repeated experiments for HeLa cells incubated with GNSs is shown in Figure 5.2 (c). In

these six IV-OCT cross sectional images, bright spots and less bright spots can be clearly seen,

giving no significant differences in the spot intensity distribution, thus demonstrating that the data

obtained in these experiments are reproducible. For the solution containing only GNSs, a single

distribution of spot intensities was observed around a central spot intensity of about 65 uarb (in the

contrast units given by the IV-OCT software, hereafter uarb) with a full width at half maximum

(FWHM) of 50 uarb. This distribution resembled a Gaussian one, as depicted by the best-fit curve


92

displayed over the distribution. Cross-sectional images of the PBS solutions containing HeLa and

Jurkat cells also show bright OCT spots, demonstrating the ability of IV-OCT to detect individual

cells. The corresponding spot diagrams are shown in the right panel of Figure 5.2. In both cases,

Gaussian distributions with peaks at about 59 uarb and an FWHM close to 26 uarb were obtained.

Thus, when compared to the histogram of IV-OCT signal intensities provided by colloidal GNSs,

it was clear that suspended cells (either HeLa or Jurkat) produce much narrower distributions,

peaking at slightly different OCT intensities. Figure 5.2 (b) also includes the intensity histograms

corresponding to the IV-OCT cross-sectional images obtained for cells that were incubated with

GNSs for 24 h. It can be clearly observed that the incubation process results in the appearance of

a number of OCT spots at higher intensities than those obtained for single GNSs and for

nonincubated cells. In these cases, the intensity histogram can be effectively described by the

superposition of two Gaussian distributions as follows: 1) the dominant contribution corresponds

to a Gaussian distribution centered at 62 uarb with an FWHM close to 25 uarb. As the distribution

peak is located at 65 uarb for free GNSs and at 59 uarb for cells only, we can consider that this

contribution is associated to free GNSs, cells that did not incorporate GNSs, and cells incubated

with several GNSs. A possible explanation of these results will be given when describing the dark

field images results. 2) The experimental data reveal an additional Gaussian distribution with a

peak at about 125 uarb and a FWHM of 50 and 26 uarb for HeLa and Jurkat cells, repectively. This

second Gaussian distribution is associated with cells that internalized GNSs, or GNSs that are

adhered to the cells during the incubation procedure. Note that in these cases, the backscattering

signal is expected to be higher, since it represents two different contributions, i.e., the

backscattered signals produced by the medium–cell interface and by intracellular GNSs.

1.2. Additional confirmation of the internalization of the GNSs into the cells

Dark field microscopy (DFM), which collects the scattered light produced by an specimen, is a

good technique to localize and monitor good light scatterers, like plasmonic nanoparticles.187

Therefore, the DFM was used to confirm the intracellular incorporation of GNSs, as shown in

Figure 5.3. Figure 5.3 (a) and (c) show, as relevant examples, DFM images of HeLa and Jurkat

cells obtained after 24 h of incubation with GNSs in PBS at of 2.5 and 1 µg/mL, respectively. In

this case, the DFM images were taken by uEye camera (Germany). According to previous studies

based on DFM visible imaging, the color of GNPs in the DFM is dependent on the scattering


93

spectra of the GNPs. For example, gold nanoparticles with scattering spectrum peaking at 547 nm

show green color in DFM, while the GNRs with scattering spectrum peaking at 687 nm show a

red color in the DFM.188 However as GNSs display a stronger scattering intensity in the infrared

region, they might not be properly imaged by a visible camera. Fortunately, due to the broad

scattering spectra of GNSs from 600 to 1400 nm (as discussed in Chapter 4), a visible camera can

still be used to image the GNSs, and the color is expected to be red. As observed in the DFM

images in Figure 5.3 (a) and (c), the red spots are thus associated with GNSs. Therefore, DFM

images provide solid evidence for the incorporation of GNSs in both HeLa and Jurkat cells. In

addition, the visualization of cells incubated with GNPs by TEM have been widely used,189, 190 due

to the different contrast and morphology that cells and GNPs show in TEM. Therefore, TEM was

used to further support the conclusions obtained by DFM. Figure 5.3 (b) and (d) show

representative TEM images of HeLa and Jurkat cells obtained after incubation with GNSs, clearly

showing the presence of GNSs in the cytoplasm of the cells.

Figure 5.3 Dark-field microscopy and bright field microscopy images (merged) of HeLa (a) and Jurkat

cells (c) incubated in a dispersion of GNSs in PBS at 2.5 and 1 µg/mL, respectively. TEM image of a HeLa

(b) and a Jurkat cell (d) incubated in a solution of GNSs in PBS at 2.5 and 1 µg/mL, respectively. The

intracellular incorporation of GNSs in cells is corroborated.


94

1.3. Incubation efficiency of GNSs incubated into the cells

The data summarized in Figure 5.2 clearly demonstrate how the incubation of the GNSs in cells

leads to complex bright spots distributions in the cross-sectional OCT images. It has been shown

that a Gaussian fitting can be used to discriminate the cells that has been efficiently incorporated

with GNSs. In order to see how the internalization efficiency was affected by the concentration of

the GNSs, IV-OCT cross-sectional images of cells incubated for 24h with GNSs at various

concentrations (from 0.05 µg/mL to 5 µg/mL) were analyzed. Figure 5.4 (a) shows the cross-

sectional IV-OCT images obtained for HeLa cells incubated with 0.05, 0.1, 0.5, 2.5 and 5 µg/mL

of GNSs. The IV-OCT image of control (cells not incubated with nanoparticles) is also shown in

Figure 5.4 (a) for comparison. With the increase of the GNSs concentration, more spots appeared

in the cross-sectional images and more bright spots were observed. The spot intensity distributions

obtained for different incubated concentrations are displayed in Figure 5.4 (b). It was previously

demonstrated that the cells that have internalized GNSs give signals in the 120-190 uarb range (red

Gauss fitting as shown in the Figure 5.2 (b)). Therefore, by direct computation of the total number

of spots with intensities in the 120–190 uarb range, it is possible to obtain a relative measure of the

population of cells with GNSs in their cytoplasm. The results obtained from the analysis of the

intensity histograms in Figure 5.4 (b) are included in Figure 5.4 (c). This figure demonstrates that

the sub-population of HeLa cells that internalizes GNSs increases as the concentration of GNSs

increases in the culture medium. IV-OCT images revealed that for GNS concentrations below 0.5

µg/mL the internalization efficiency strongly depends on the GNS concentration, whereas for

larger concentrations, the internalization efficiency tends to saturated and so is only slightly

dependent on the incubation concentration. Interestingly, the number of incubated cells (>120 uarb)

reaches a maximum for the highest concentration of GNSs (2656) that is quite similar to the same

number of HeLa cells (2735) in the control. Therefore, it can be considered that OCT can be

efficiently used for imaging of HeLa cells.


95

Figure 5.4 (a) IV-OCT cross-sectional images of nonincubated HeLa cells and HeLa cells incubated for

24 h with various concentrations of GNSs. (b) Intensity histograms as calculated from the IV-OCT cross-

sectional images for spots intensities larger than 120 uarb. (c) Number of cells with intensity above 120 uarb

generated from HeLa cells incubated with various concentrations of GNSs for a fixed time of 24 h.

A similar trend was observed when the internalization efficiency of GNSs by Jurkat cells was

determined based on IV-OCT images, as shown in Figure 5.5 (a). The spot intensity distributions

obtained for various incubated concentrations are displayed in Figure 5.5 (b), where the intensity


96

range was restricted to 100-160 uarb. Interestingly, in the case of Jurkat cells, which present a lower

phagocytosis ability than HeLa cells, the GNSs were incorporated in the cells in a sufficient

amount to increase the OCT signal, enabling the discrimination between individual cells and cells

with GNSs. The integrated OCT intensity with spot intensity above 100 as function of GNSs

concentrations is shown in Figure 5.5 (c). The integrated OCT intensity shows also a similar trend

with increase of GNSs concentrations to that observed for incubated HeLa cells (see Figure 5.4

(c)).


97

Figure 5.5 (a) IV-OCT cross-sectional images of nonincubated Jurkat cells and Jurkat cells incubated for

24 h with various concentrations of GNSs. (b) Intensity histograms as calculated from the IV-OCT cross-

sectional images for spots intensities larger than 100 uarb (associated only with signals of cells that have

incorporated GNSs). (c) Integrated OCT intensities generated from Jurkat cells incubated with various

concentrations of GNSs for a fixed time of 24 h. The OCT signal was calculated by integrating the

frequency distribution signals for spot intensities above a threshold intensity of 100 uarb (IV-OCT signal of

successfully incubated cells).

Figure 5.6 Dark-field microscopy images of HeLa cells incubated with various concentrations of GNSs.

Incubation time was 24 h in all cases: The scale bar is 20 µm.

A comparison of data obtained by IV-OCT and those obtained by DFM was performed for HeLa

cells. The DFM images obtained for HeLa cells incubated with various GNSs concentrations were

systematically studied, representative images are shown in Figure 5.6. These images clearly

indicate that the number of red spots in cells increases with the concentration of GNSs in the

incubation medium. The analysis of DFM images also allows for a qualitative evaluation of the

adhesion efficiency of GNSs, by computing the average number of “red spots” per cell as a

function of the GNS concentration. Assuming that each “red spot” corresponds to an individual


98

GNSs, it can be discussed how the number of GNSs incorporated to cells changes with the GNSs

concentration as shown in Figure 5.6. The concentration-dependent incubation efficiencies,

determined by DFM, for HeLa cells are included in Figure 5.7, together with the OCT trends

obtained from Figure 5.4 (b) (i.e. the spots with OCT spot intensity above 120 uarb). The average

number of GNSs in cells were calculated by counting the red spots in the HeLa cells, using the

images shown in the Figure 5.6 It is clear that the DFM results were consistent with the IV-OCT

results. The DFM images also reveal that the incorporation (or adhesion) efficiency of GNSs in

living cells increases monotonically as the concentration of the GNSs in the culture medium

increases. It can be also observed a tendency towards saturation for concentrations larger than 0.5

µg/mL, in very good agreement with the data extracted from IV-OCT images. This saturation

behavior, evidenced by both DFM and IV-OCT, was in accordance with previous studies showing

that the cellular uptake efficiency displays a saturation behavior as the concentration of

nanoparticles increases.51, 191 This is due to the limited number of available serum proteins in cells,

which plays an important role in the internalization of gold nanoparticles.191 In the DFM image of

HeLa cells incubated with 0.05 µg/mL of GNSs, there are only 2 GNSs per cells (in average). By

observing the corresponding OCT signal (Figure 5.4 and Figure 5.7), there are no spots with

intensity above 120 uarb. Thus, HeLa cells incubated with 2 GNSs do not lead a contrast

enhancement in the OCT cell imaging; the signal must be so low that it is impossible to detect the

HeLa cells incorporated with GNSs. For HeLa cells incubated with 5 µg/mL of GNSs, the spots

outside the cells are due to GNSs that have not been removed by the washing process. Note that

some HeLa cells do not have as many GNSs as other HeLa cells, thus they may lead to the different

OCT intensity in the OCT cross-sectional. For example, HeLa cells incubated with 1 or 2 GNSs

result in a low OCT intensity (below 120 uarb), and HeLa cells incubated with 10 to 45 GNSs

produce a high OCT intensity (above 120 uarb). This fact may give a reasonable explanation to the

blue Gauss fitting in Figure 5.2 (b) with OCT intensity distribution from 40 to 120 uarb. This OCT

intensity distribution must be due to the backscattered light produced by the free GNSs,

nonincubated HeLa cells, and HeLa cells incubated with several GNSs.


99

Figure 5.7 Average OCT spot intensities and number of GNSs in cells generated from HeLa cells incubated

with various concentrations of GNSs for a fixed time of 24 h. The average OCT spot signal was calculated

from the frequency distribution signals for spot intensities above a threshold intensity of 120 uarb (IV-OCT

signal of successfully incubated cells), the number of GNSs in cells were calculated from DFM images.

1.4. Contrast enhancement at cellular and tissue level

Finally, it should be noted that the unequivocal relation between OCT spots with intensities

larger than 120 uarb and GNS internalization in cells allows for not only detailed studies of the

incubation efficiency, but also for the three-dimensional localization of cells with GNSs in their

cytoplasm. An example of this imaging capacity is given in Figure 5.8. In Figure 5.8 (a), the raw

IV-OCT image of a suspension of HeLa cells after incubation for 24 h with 2.5 µg/mL of GNSs is

shown. The IV-OCT raw image shows a large density of spots with a wide variety of intensities

ranging from 40 to 180 uarb. As discussed above, these spots reveal a high density of GNSs,

suspended HeLa cells and suspended HeLa cells that have internalized GNSs. In order to isolate

the images of HeLa cells that have internalized GNSs efficiently, the raw IV-OCT images were

treated by filtering the spots above a single threshold level. Figure 5.8 (b) shows filtered the IV-

OCT image corresponding to the data shown in Figure 5.8 (a) after removing the OCT spots with

intensities below 120 uarb. Since OCT intensities above 120 uarb are only generated by HeLa cells


100

with internalized GNSs, Figure 5.8 (b) corresponds to an image of the sub-population of HeLa

cells containing GNSs (they have been labeled in green color). Therefore, IV-OCT could be used

for molecular imaging of internalized cells. As explained in the Experimental techniques section,

the clinical IV-OCT system could also be used for three-dimensional imaging by pulling back the

catheter. This possibility, together with the threshold level filtering, was exploited to obtain a three-

dimensional image of the sub-population of HeLa cells with internalized GNSs. The pull back

video was imported to ImageJ program, after setting a threshold level filtering, HeLa cells with

internalized GNSs were distinguished, and then their coordinates (x, y, z) were saved and imported

to the Origin program. The three dimensional image was drawn by means of 3D scatter program.

Thus, the three dimensional image of the sub-population of HeLa cells with internalized GNSs is

shown in Figure 5.8 (c). The red spots correspond to HeLa cells with internalized GNSs, the

coordinates x and y correspond to the position of HeLa cells in the OCT cross-sectional image, and

z coordinate is obtained from the OCT video (540 slices correspond to 54 mm of the pull back

distance). This result indicates the possibility of real-time, three-dimensional tracking of living

cells in the intravascular system by employing specifically targeted GNSs.

Figure 5.8 (a) Raw cross-sectional IV-OCT image of a suspension of HeLa cells after incubation with

GNSs for 24 h. The concentration of GNSs in the culture medium during incubation was 2.5 µg/mL. (b)

IV-OCT cross-sectional image of the same cell suspension used in (a), but after applying a threshold

filtering to the raw image. The intensity threshold was set to 120 uarb, so that observed spots correspond to

HeLa cells that internalized GNSs. (c) Three-dimensional IV-OCT image of the suspended HeLa cells that

internalized GNSs. Each pixel providing a IV-OCT signal larger than 120 uarb was identified as a single

HeLa cell with internalized GNSs. Scale bar represents 1 mm.

OCT Catheter

OCT Catheter

Tubing Tubing (a) (b) (c)


101

As mentioned above, many clinical applications also require contrast enhancement within

tissues. In order to demonstrate the potential application of GNSs to enhance the OCT contrast

inside a tissue, a final experiment was designed in which the IV-OCT catheter was positioned

inside a muscular tissue (Gallus gallus domesticus). Then, two injections (0.5 mL per injection) of

an aqueous solution of GNSs (50 µg/mL) were performed in the proximity of the OCT catheter

(see Figure 5.9 (a)). The cross-sectional images of the muscular tissue obtained before and after

the injection of GNSs are shown in Figure 5.9 (a) and (b). The circle in the center of the image

corresponds to the catheter. The red yellow area around the IV-OCT catheter is due to the

backscattered light produced by the tissue and the GNSs. According to the optical scattering

properties of soft tissues reported by Joseph,192 tissues show a high scattering coefficient from 600

to 1400 nm, and so giving a high OCT intensity around the catheter. In addition, the OCT intensity

decreases from the OCT catheter to deeper tissue. A detailed comparison of these figures reveals

a contrast enhancement at the injection locations. This is clearly manifested in Figure 5.9 (c),

which includes the average OCT intensity (IOCT) against depth profiles obtained along the scan

direction represented by the green arrows in Figure 5.9 (a) and (b), i.e., along two scan directions

crossing the GNS injection areas. Figure 5.9 (c) clearly demonstrates that OCT intensity decrease

as function of the depth due to the scattering and absorption properties of the tissue. In addition,

the presence of GNSs leads to an OCT contrast enhancement between 0.8 and 1.6 mm, in good

agreement with the location of the injected GNSs. This result is more clearly observed when the

OCT depth profile obtained prior to the injection is subtracted from the profile obtained after

injection (IOCT), shown in Figure 5.9 (d). This figure shows the relevant enhancement in the OCT

contrast produced by the GNSs inside the tissue. Indeed the distribution of GNSs as function of

depth can be observed. Thus, the data included in Figure 5.9 clearly demonstrate the potential use

of GNSs as contrast agents for OCT imaging of tissues using cardiovascular clinical systems.


102

Figure 5.9 (a) Cross-sectional OCT image of a tissue prior to the injection of GNSs. An aqueous solution

of GNSs (0.5 mL; 50 µg/mL) was injected in the areas delimited by the blue dashed lines. (b) Cross-

sectional OCT image of the same tissue shown in (a) after the injection of an aqueous solution of GNSs.

(c) OCT intensity (IOCT) versus depth in the tissue, as obtained along the scan lines, is indicated by the green

arrows in (a) and (b). (d) Difference between the OCT intensity obtained prior to and after the injection of

GNSs (IOCT) as a function of the penetration depth in the tissue. The spatial extension of the GNSs within

the tissue is indicated by the gray areas included in (c) and (d).

1.5. Internalization of GNSs in HMEC-1 cells under flow conditions

HMEC-1 is a human microvascular endothelial cell line, and the dysfunction of this cell line

related to cardiovascular disease. Evaluation of internalization of GNSs in HMEC-1 under flow

conditions is a very important step for clinical application of GNSs in cardiovascular disease.

Therefore, I have settled a flow setup as described in Experimental techniques. Due to the limited

time of project, the functionalization of GNSs is not available. In this case, commercial GNSs

coated with PEG without functionalization and HMEC-1 without activated were used to evaluate

the flow system. The DFM images (taken by EMCCD camera) of GNSs and HEMC-1 at different


103

flow times are shown in Figure 5.10. The morphology of HEMC-1 can be clearly seen. The

movement of GNSs 1 and 2 can be tracked as labeled by red arrows. The internalization of GNSs

in HEMC-1 haven’t been found due to the non-specific functionalized of GNSs and non-activated

HEMC-1. More work needed to be done in order to see in the internalization of GNSs in HEMC-

1 under flow condition.

Figure 5.10 DFM images of GNSs and HMEC-1 cells at different flow time (0 s and 1 s). The positions of

GNSs 1 and 2 at different flow time were labeled by red arrows.

1.6. Conclusions

In summary, in this chapter, I have shown how clinical IV-OCT allows for the detection of

individual cells suspended in biological fluid. The experimental results indicated that the

backscattering contrast of individual cells could be substantially enhanced by the intracellular

incorporation of specifically designed GNSs with high scattering cross-sections at the IV-OCT

operating wavelength (1320 nm). A detailed intensity analysis of IV-OCT images enabled the

identification and discrimination of different sub-populations within a suspension of living cells.

In particular, an analysis of the intensity histograms from IV-OCT images allowed for the detection

of living cells in which GNSs attachment occurred, and made it possible to estimate the attachment

efficiency and its relationship to the concentration of GNSs in the culture medium. These results

were in excellent agreement with those obtained using alternative techniques, such as DFM and


104

TEM. Finally, the tissue contrast enhancement by OCT, achieved after the injection of GNSs inside

the tissue was demonstrated.

As GNSs are easy to functionalize with biomolecules to target specific cell sites, the results of

this study provide a new path for reliable molecular imaging by IV-OCT, and thus constitute the

first step towards the development of new diagnostic procedures at the clinical level.

Chapter 6. PbS QDs as

contrast agents for

cardiovascular bimodal

imaging

Chapter 6. PbS QDs as contrast agents for cardiovascular bimodal imaging

106

6.1. Introduction

Up to now, silica-coated gold nano-shells with plasmon resonance close to the IV-OCT laser

wavelength (1.3 µm) and high back-scattering cross sections were demonstrated as the best nano-

contrast agents, showing significant IV-OCT contrast enhancements in suspensions and tissues.179

In addition, these nanoparticles have been used for OCT molecular imaging of several cell lines.193

The potential application of OCT contrast agents would further benefit if the NPs, in addition to

the OCT contrast enhancement, could provide simultaneous photoluminescence (PL) contrast.

This would open the possibility of achieving multimodal intracoronary imaging that would be

immensely helpful, for instance, in the understanding of the relation between atherosclerotic

plaque microstructures and the molecular mechanisms that underlie their formation and instability,

leading to acute coronary syndromes.194 A pioneering work of H. Yoo and co-workers60

demonstrated that the simultaneous acquisition of multimodal OCT+(PL) intracoronary images

allowed for microstructural and molecular functional image, an essential information to the

correlation of coronary artery disease and vessel wall healing. These authors used the Cy7

fluorophore to provide fluorescence imaging while intracoronary OCT images were constructed

from the inherent contrast provided by the artery.195, 196 However, this approach implied the use of

two laser sources (one to excite the fluorophore and the other for OCT),60 increasing the

complexity of the experimental arrangement. Solving this drawback implies the use of a contrast

agent capable of providing both luminescence and back-scattering contrast just under single laser

excitation at 1.3 µm. Furthermore, if a NP is to be used as a multimodal OCT+PL agent in

cardiological applications, its luminescence should be produced within the so-called infrared

biological windows,51 where human tissues become partially transparent allowing high penetration

in in vivo fluorescence imaging.197-199 In, addition, NPs is much more stable than the fluorophores

and doesn’t have photobleach.53


107

Figure 6.1 Schematic representation of a NP capable of OCT+PL imaging under single laser line excitation

at 1.3 µm.

In this chapter, the potential use of Infrared emitting Quantum Dots (IR-QDs) as OCT+PL

multimodal intracoronary imaging agents under single line laser excitation was evaluated. The

operating scheme is shown in Figure 6.1. Thus under OCT laser excitation, the IR-QDs can scatter

the OCT laser light, the backscatter light can be collected by the IV-OCT catheter, and thus

produces OCT signal. In addition, the IR emission of QDs can be generated under OCT laser

excitation, and thus the fluorescence image produced by the IR-QDs can be obtained by using a

near infrared camera. The simultaneous capacity of these dots for back-scattering and fluorescence

generation in III-BW52 under single 1.3 µm laser excitation was evaluated in aqueous suspensions,

tissues and ex vivo aorta by using an IV-OCT clinical equipment.

6.2. Characterization of PbS QDs

The IR-QDs used all along this work were PbS QDs (CANdots® Series C with a nominal

emission wavelength of 1600 nm), and they were surface-functionalized with a PEG-ligand

(DSPE-PEG-amine) in order to achieve dispersability in water. A carton of PbS QDs with PEG is

drawn in Figure 6.2. Details of surface functionalization can be found in the Experimental

techniques section. A representative TEM image of the IR-QDs is shown as Figure 6.2 (a).

Statistical analysis of this and other TEM images reveals an average IR-QD size of 5.4 nm, with a

size dispersion of ±2.3 nm, as can be appreciated from the size histogram included as an inset in


108

Figure 6.2 (a). Therefore, taking into account the density of PbS (7.6 mg/cm3) and the volume of

each PbS QDs is (4/3)×π×(2.7×10-7)3 cm3, the number of QDs per mL for initial content (0.6

mg/mL) is 9.21×1017 NPs/mL. Figure 6.2 (b) shows the room temperature extinction cross section

together with the emission spectrum of the IR-QDs. The extinction cross section clearly evidences

the first exciton absorption peak at around 1420 nm. The non-negligible extinction at the OCT

laser wavelength peak (also included in this figure) should also be noted. The OCT laser spectrum

matches the extinction spectrum, and therefore, the PbS QDs can be excited by using a IV-OCT

laser. The emission spectrum of IR-QDs is constituted by a single broad band centered at around

1.55 µm (III-BW), so that it can be easily spectrally isolated from the IV-OCT laser radiation by

using long pass filters.

Figure 6.2 (a) Characteristic TEM image of the QDs used in this work. Inset shows the size histogram

revealing an average size close to 5 nm. (b) Blue line corresponds to the extinction cross section of the QDs

(Commercial PbS QDs without surface modification, the concentration of PbS is 10 mg/mL). The emission

spectrum of the QDs is also included (red line). The orange line corresponds to the spectral distribution of

the laser radiation generated by the IV-OCT catheter.


109

6.3. Bimodal OCT and fluorescence imaging of colloidal QDs

Figure 6.3 Schematic representation of the experimental approaches used to test the capacity of IR-QDs

for multimodal OCT+PL imaging in a fluid.

Figure 6.4 (a) Cross-sectional OCT images of the tubing in the presence of three aqueous solutions of IR-

QDs, GNRs and GNSs. The cross sectional image in absence of any NPs is also included as “control image”.

Scale bar, 1 mm. (b) Integrated OCT intensity normalized by the total mass of NPs as obtained for IR-QDs,

GNRs and GNSs. (c) Integrated OCT intensity normalized by the total number of NPs as obtained for IR-

QDs, GNRs and GNSs.


110

The experimental arrangement depicted in Figure 6.3 is used to evaluate the bimodal OCT+PL

imaging of colloidal QDs introduced in a tubing (simulating an artery). The IR-QDs were injected

in the tubing in which contains the IV-OCT catheter, the static OCT was operated to evaluate the

OCT contrast produced by the IR-QDs. Figure 6.4 (a), left side, shows the IV-OCT cross sectional

image when the tubing was filled with a clean suspension of water, which was used as a control

image for IV-OCT experiments. The cross sectional IV-OCT image obtained when the tubing was

filled with an aqueous suspension of IR-QDs (0.06 mg/mL) is also included in Figure 6.4 (a). The

contrast enhancement produced by the suspended IR-QDs is clearly evidenced. This fact

unequivocally reveals the capacity of individual IR-QDs (in spite of their small sizes) to

backscatter the 1.3 µm OCT laser radiation. The scattering properties of QDs in the visible domain

were already theoretically postulated and experimentally evidenced by dark field microscopy

previously.200 Nevertheless, this is the first time that such scattering ability is demonstrated in the

infrared spectral region, and particularly under the IV-OCT excitation conditions. In order to

compare the IV-OCT enhancement produced by IR-QDs in respect to that provided by metallic

nanoparticles, Figure 6.4 (a) also includes cross sectional images obtained when the tubing was

filled with an aqueous suspensions of GNPs. Two types of GNPs have been used here for

comparison. GNRs (100 nm length and 10 nm width) with a plasmon resonance wavelength peak

close to 1300 nm and GNSs (198 nm core and 16 nm shell) with a multimodal plasmonic band

extending from 800 up to 1800 nm. Concentrations of the solutions of both GNRs and GNSs were

0.06 mg/mL. Figure 6.4 (b) shows the IV-OCT signal enhancement (in respect to the control)

normalized by the mass of contrast agents, as obtained for the three particles investigated. Figure

6.4 (b) reveals that, for similar concentrations, IR-QDs show an IV-OCT contrast per mass unit

enhancement smaller but comparable to that provided by GNPs. Figure 6.4 (c) shows the IV-OCT

integrated intensity normalized to the number of particles. In this case, it is clear that the IV-OCT

intensity generated by a single metallic nanoparticle is more than eight orders of magnitude larger

than that provided by individual IR-QDs. Indeed, this was expected because of two main reasons.

The first one is the large size difference: the volume of an individual GNS is more than four orders

of magnitude larger than the volume of an individual IR-QDs. Secondly, for metallic nanoparticles

the back-scattering is associated with the collective resonant excitations of charges (plasmon

resonance overlaps with the IV-OCT laser radiation).148, 201 As a consequence, the back-scattering

cross section of, for instance, GNSs with enough plasmon resonance at OCT laser wavelength are


111

as large as 7×10-11 cm2 per nanoparticle.179 The scattering cross section of IR-QDs has not been

reported up to now, but theoretical studies predict that the scattering cross section of individual

QDs is much smaller than the absorption cross section.202 This applies when the QD’s size is much

smaller than the wavelength of the incident light, as it occurs in this work. The extinction cross

section at 1.3 µm of our IR-QDs has been estimated to be close to 2.5×10-19 cm2.203 This is, indeed,

eight orders of magnitude smaller than the scattering cross section of GNSs, in good accordance

with the eight orders of magnitude differences experimentally obtained between the IV-OCT

intensity provided by individual IR-QDs (1.8×10-16) and individual GNSs (5.4×10-8) (see Figure

6.4 (c)). Despite the reduced IV-OCT signal, it is important to remark that the contrast generated

by individual IR-QDs, although much weaker than that provided by individual metallic

nanoparticles, is still well measurable by the clinical IV-OCT system.21

Figure 6.5 (a) Cross sectional OCT and infrared as obtained for different concentrations of IR-QDs in the

tubing. Scale bar, 1 mm. (b) Integrated OCT intensity as a function of the concentration of IR-QDs. Scale

bar, 4 mm. (c) Integrated infrared luminescence intensity as a function of the concentration of IR-QDs. In

(b) and (c) dots are experimental data and solid line is a guide for the eyes.

Once the capacity of IR-QDs for providing IV-OCT contrast has been demonstrated, their ability

to simultaneously produce PL contrast was evaluated by using an infrared fluorescence camera to

image the tubing (see Figure 6.3). The infrared fluorescence camera was set above the tubing to

take the infrared image, meanwhile, a long pass filter 1350 nm was put before the camera to


112

remove the OCT laser signal. The obtained infrared fluorescence images of a tubing filled with

aqueous suspensions of IR-QDs at different concentrations are shown in Figure 6.5 (a), from

which the presence of IR-QDs is evidenced. This indicates efficient absorption and re-emission of

the 1.3 µm OCT laser radiation by IR-QDs. At this point, it is important to remark that the data

included in Figure 6.4 and Figure 6.5 (a) clearly demonstrate the ability of our IR-QDs to produce,

simultaneously, IV-OCT and IR luminescence images, all under just the single laser line excitation

of the cardiovascular OCT system. The dependences of both integrated OCT and luminescence

intensities (𝐼𝑂𝐶𝑇 and 𝐼𝑙𝑢𝑚, respectively) on the concentration of IR-QDs are shown in Figure 6.5

(b) and (c). In both cases the integrated intensities increase monotonously with the IR-QD

concentration, but with very different trends. On one hand, the IV-OCT intensity vs the IR-QDs’

concentration shows a linear dependence in a Log-Log representation. This implies that the IV-

OCT intensity (𝐼𝑂𝐶𝑇) can be written as 𝐼𝑂𝐶𝑇 = [𝑄𝐷𝑠]𝑛, where [𝑄𝐷𝑠] is the concentration of IR-

QDs. Linear fitting of experimental data in the Log-Log representation provides, for this particular

case, a value of 𝑛 = 0.3, so that the IV-OCT intensity increases with the concentration of IR-QDs

following a sublinear trend. The reason explaining this sublinear trend is not clear at this point.

Nevertheless it should be mentioned here that a similar trend was also observed for metallic

nanoparticles,179 so that it is likely that the sublinearity is related to the way in which the processing

and quantification of IV-OCT images are performed by the clinical equipment. On the other hand,

the luminescence intensity generated by IR-QDs (when optically excited with the IV-OCT laser)

was found to follow a linear trend with the concentration of IR-QDs (see Figure 6.5 (c)), as it was

indeed expected. This opens the door to straightforward determination and quantification of

intracoronary accumulation of QDs by just monitoring the infrared luminescence generated from

the analyzed arteria under excitation with the IV-OCT catheter.


113

Figure 6.6 (a) Infrared fluorescence images of a catheter filled with an IR-QD suspension (0.6 mg/mL) as

obtained for different tissue thickness. Scale bar, 4 mm. (b) Integrated infrared luminescence intensity as a

function of the tissue thickness. Dots are experimental data obtained from the luminescence images in (a)

and the red line is the best fit to an exponential decay. The penetration length provided by exponential fit

is 0.8 mm.

The potential application of infrared fluorescence imaging for sub-tissue localization of QDs in

arteries would be determined by the effective penetration length (under excitation with the IV-

OCT catheter). In order to evaluate this important aspect, different thickness of the tissues varing

from 0 up to 6 mm, were set above the tubing. Then the fluorescence images generated by the IV-

OCT catheter from a tubing filled a QDs/water solution were obtained for different tissue thickness

by the IR camera as shown in Figure 6.6 (a), the fluorescence intensity decreases when increasing

the tissue thickness, and the fluorescence image is almost invisible when the tissue thickness

reached 3mm. The integrated fluorescence intensity (𝐼𝑙𝑢𝑚) as a function of the tissue thickness (𝑑)

is shown in Figure 6.6 (b). According to previous stablished models of light propagation in tissues,

the fluorescence intensity should follow a typical trend given by:

𝐼𝑙𝑢𝑚 = 𝐼0𝑒𝑥𝑝[− 𝑑 𝑙𝑝⁄ ] (6.1)

where 𝐼0 is a constant and 𝑙𝑝 is the so-called optical penetration length. Experimental data fit well

to this trend, providing an optical penetration length close to 0.8 mm. Indeed, this is quite close to

the optical penetration lengths reported for different infrared emitting luminescence probes that


114

have been successfully used for high penetration and high resolution “in vivo” imaging.204-206 Note

that the maximum depth at which IR-QDs can be detected depends on the background level of the

IR camera. In comparison to our camera, working at 0 ºC, the use of an IR camera working at -80

ºC (operating temperature of double Peltier cooled commercial cameras) would lead to a reduction

in the background signal of more than one order of magnitude, and so making possible to detect

IR-QDs at subcutaneous depths well in excess of 1 cm.

At this point it should be mentioned that future optimization of the fluorescence and the

scattering efficiencies of a nanostructure for bimodal OCT+PL imaging is not an easy task as they

are exclusive processes. In fact, if a nanoparticle shows high scattering cross sections, then the

amount of absorbed light (that determines the fluorescence intensity) will be reduced. Thus, a

compromise has to be reached. Such compromise could be obtained in the future by using hybrid

nanostructures combining, into a single unit, highly scattering units and bright fluorescent units.

This could also be achieved through core/shell engineering (by creating a thin metallic shell on the

surface of IR-QDs) or, alternatively, by linking IR-QDs and metallic nanoparticles through surface

modification. In addition, I would also like to note that, with the experimental data available now,

I cannot disregard the contribution of absorption-reemission of pump photons to the overall back-

scattering signal. Nevertheless, I consider that it would be a second order process when compared

with the direct backscattering due to photon backscattering at the surface of IR-QDs. Furthermore,

even in the case that these processes were taking place, they would not contribute to the observed

OCT contrast due to the lack of coherence between pump and backscattered light.

6.4. OCT and fluorescence bimodal imaging in tissues

As commented in the introduction, one of the advantages of IV-OCT is its capacity to provide,

in addition to a topographical image of an artery, information about tissue properties beyond the

artery wall. Thus, a multimodal OCT+IR probe should be able not only to provide dual contrast

when being in suspension (inside the artery) but it should also be capable of dual contrast when

being located inside a tissue.60, 106, 207 The ability of our IR-QDs for OCT+PL dual contrast imaging

in tissues was, therefore, also explored.


115

Figure 6.7 Schematic representation of the experimental approaches used to test the capacity of IR-QDs

for multimodal OCT+PL imaging in a fluid.

The experimental approach for dual contrast imaging in tissue is schematically drawn in Figure

6.7. Instead of using the tubing (see Figure 6.3), a muscular tissue was used. The IR-QDs were

injected in the tissue but near the IV-OCT catheter in order to detect the OCT contrast enhancement

in tissue produced by the IR-QDs and at the same time, making an efficient excitation of the IR-

QDs by the OCT laser. A cross sectional IV-OCT image of the tissue before the QDs’ injection is

shown in Figure 6.8 (a). It is constituted by a diffuse background, associated to the characteristic

inhomogeneous refractive index of tissues. Figure 6.8 (c) shows the corresponding infrared

fluorescence image of the tissue, demonstrating, as expected, no contrast at all. Indeed, this result

indicates the negligible excitation of tissue auto-fluorescence by the 1.3 µm OCT laser. When IR-

QDs were injected into the tissue both OCT and luminescence images change appreciably in the

injection area in respect to the control (no IR-QDs). Figure 6.8 (b) shows the cross sectional OCT

image of the tissue after the injection of 200 µL of an aqueous suspension of IR-QDs (0.6 mg/mL).

The location of the injected IR-QDs is indicated by a red arrow. The presence of IR-QDs in the

tissue is observed by a clear increment in the OCT intensity. This is further evidenced in Figure

6.8 (e) in which the OCT signal profile (obtained inside the tissue along the scan direction

represented by the yellow arrows in Figure 6.8 (a) and (b)) was included. Comparison between

the OCT intensity profiles obtained, before and after injection, clearly demonstrates a remarkable

intensity enhancement at the location of the IR-QDs. Indeed, the OCT signal was increased by


116

almost 2.5 times due to the IR-QD induced backscattering. At the same time, the presence of IR-

QDs was simultaneously monitored by their infrared luminescence, as it is demonstrated in the

infrared fluorescence image included as Figure 6.8 (d). Thus, the ability of IR-QDs for dual

OCT+PL and fluorescence imaging inside tissues is clearly proved. This multimodal imaging

capacity appears of special relevance for cardiovascular imaging, as it would allow for unequivocal

identification of inner lesions beyond the artery contour by using adequately functionalized IR-

QDs.

Figure 6.8 (a) Cross-sectional OCT image of a tissue before the injection of IR-QDs. Scale bar, 1 mm. (b)

Cross-sectional image of the same tissue after the injection of IR-QDs. The location of the IR-QDs is

indicated by the red arrow. The dashed red line indicates the position of the tissue-air interface. (c) Infrared

fluorescence image of the tissue prior to the injection of IR-QDs. (d) Infrared fluorescence image of the

tissue after the injection of IR-QDs. Scale bar, 4 mm. (e) OCT intensity profile as obtained along the yellow

arrows indicated in (a) and (b). Data obtained before and after injection are included denoting the

enhancement in the OCT signal due to the presence of IR-QDs.


117

6.5. OCT and fluorescence bimodal imaging in rabbit artery

Figure 6.9 shows an optical picture of the experimental set-up used for multimodal OCT+PL imaging of

an artery excised from a rabbit. The different parts are indicated and labelled.

In order to finally test the potential application of our IR-QDs for multimodal cardiovascular

imaging, ex vivo experiments were conducted. The aorta of a rabbit (whose dimensions are similar

to human coronary arteries) was excised from a sacrificed rabbit. The branches were sealed by

suture procedures and the IV-OCT catheter was introduced in the artery by using a cannula and

metallic guide. Details about the experimental procedure can be found in Figure 6.9. The IR-QDs

were injected in the artery by using a syringe, which connects to the cannula by a tubing. The IV-

OCT catheter was introduced and was scanned along the artery with a speed of 10 mm/s. During

the scanning of the IV-OCT catheter the fluorescence generated by the IR-QDs in the artery was

detected by the same InGaAs infrared camera used in previous experiments. This simple

experimental set-up made possible the simultaneous acquisition of the OCT and infrared

fluorescence images of the artery in real time.


118

Figure 6.10 Optical image of the aorta excised from a sacrificed rabbit. The position of the IV-OCT catheter

during axial scanning is indicated by red arrow. The instantaneous (real time) and accumulative infrared

fluorescence image generated by the IR-QDs inside the artery are also shown for the different positions of

the IR-OCT catheter and named A, B, C, D and E. The corresponding OCT cross sectional images are also

displayed together with the transverse OCT image of the artery. The shadow in OCT cross section images

are due to the light blocked by the metallic guide. The scale bar for optical and fluorescence image is 4 mm,

for OCT cross section image is 1 mm.

Figure 6.10 shows the optical image of the rabbit aorta at different times during the IV-OCT

scanning. The position of the OCT catheter at each time, which it is indicated by an arrow, can be

elucidated thanks to a red laser (at 650 nm) that is integrated into the rotating fiber. The real time

and accumulative fluorescence images obtained at different times are also included in Figure 6.10.

In addition, for the sake of clarify, the accumulated fluorescence image are also include in this


119

figure. The fluorescence images are due to the emission of PbS QDs under excitation of IV-OCT

catheter. IV-OCT cross-sectional images of the artery were simultaneously acquired, five

representative OCT cross sectional images obtained at the different artery positions (indicated by

A, B, C, D, E arrows) are also included in Figure 6.10. The shadow in the OCT cross section

images are due to the OCT laser light blocked by the metallic guide. The OCT contrast in the artery

indicated the existence of IR-QDs in the artery. In addition, an axial OCT image of the artery,

which was obtained after the IV-OCT scanned the artery from 0 to 54 mm was included in Figure

6.10. This image shows the morphology of the artery used in this experiment, the indentations in

the artery (as indicated by A, B, C, and D arrows) are supposed to help the accumulation of the

IR-QDs in the artery. A video of the measuring process in which the translation of the IV-OCT

and the simultaneous acquisition of fluorescence image can be found online

(https://drive.google.com/file/d/0B5z1PgUYHn1eaTV1b08zUjRrMGM/view?usp=sharing).

The fluorescence images clearly demonstrate how the PL generated by the IR-QDs inside the

artery accurately reproduces its morphology. In addition, several bright spots (Indicated by A, B,

C and D arrows) are observed, revealing the accumulation of IR-QDs inside the artery. When these

images are compared with the OCT ones, it is possible to correlate these bright spots with the

localization of the arterial branches that were clamped. This is evidenced by arrows in the OCT

longitudinal image of the artery. Thus, I state that at these positions IR-QDs are accumulated. The

experiments included in Figure 6.10, do not only reveal the potential use of IR-QDs for

multimodal cardiovascular imaging, but also demonstrate that the synergy between fluorescence

and OCT cardiovascular imaging could result in the appearance of advanced imaging and

diagnosis tools.

6.6. Conclusions

In this section the capability of semiconductor quantum dots, emitting in the third biological

window, for multimodal cardiovascular imaging has been demonstrated. Infrared emitting

quantum dots were shown to be capable of providing simultaneous contrast in intracoronary optical

coherence tomography and infrared fluorescence imaging all under 1.3 µm wavelength single

beam laser excitation provided by a clinical catheter. It is demonstrated that , when a single infrared

emitting quantum dot is optically excited by a 1.3 µm wavelength photon, the back-scattered

https://drive.google.com/file/d/0B5z1PgUYHn1eaTV1b08zUjRrMGM/view?usp=sharing


120

radiation is strong enough to produce a measurable contrast by a clinical OCT system. At the same

time, the absorbed photons are efficiently re-emitted at 1.55 µm, making possible their

simultaneous localization by fluorescence imaging. Dual fluorescence and optical coherence

tomography imaging was demonstrated both in colloidal suspension and in tissues. Furthermore,

a simple “proof of concept” demonstration at the ex vivo level showed the real potential of infrared

emitting quantum dots for multimodal imaging of arteries, pointing out the synergy between

fluorescence imaging and optical coherence tomography.

Consequently, infrared emitting quantum dots appear as the first dual contrast agents for

intracoronary imaging based on the use of already established optical coherence tomography

catheters. The possibility of acquiring multimodal intracoronary imaging opens a new avenue

towards the development of new diagnostic techniques based on multimodal imaging.

Chapter 7. Conclusions and

perspectives

Chapter 7. Conclusions and perspective

122

7.1. GNSs, best OCT contrast agent

Different kinds of GNPs show different OCT contrast enhancement due to their different

scattering properties, as OCT is based on the analysis of backscattered signals. The GNSs with

size around 220 nm show a higher scattering cross section at the OCT laser wavelength (1.3 µm)

than GNSTs and GNRs, thus the GNSs investigated in my thesis are considered as the best contrast

agents for IV-OCT.

7.2. Single GNSs visualization by IV-OCT

The IV-OCT can be used to track single GNSs due to the high sensitivity of IV-OCT system

and high scattering cross section of GNSs.

7.3. GNSs can act as contrast agents for cell imaging and tissue contrast enhancement

by IV-OCT

The analysis of the intensity histogram for IV-OCT cross-sectional images of GNSs incubated

in the HeLa and Jurkat cells allows the detection of living cells with GNSs attached. The number

of detected cells with GNSs by IV-OCT as function of GNSs concentration demonstrated a

saturation trend in high GNSs concentration. The internalization of the GNSs in the cells have

been confirmed by TEM and DFM.

The injection of GNSs in the tissue near the IV-OCT catheter shows a contrast enhancement,

the OCT signal after subtract the background allow visualization of GNSs distribution in the tissue.

7.4. PbS QDs could act as a dual contrast for cardiovascular imaging

The broad absorption of PbS QDs allows the excitation of their luminescence by the OCT laser

at 1.3 µm. After injection of PbS QDs in artery, their emission around the 1.55 µm after excitation

by OCT laser could be used for fluorescence imaging of the artery in the third biological window.

In addition, the assignable scattering properties of PbS QDs could be used for OCT contrast

enhancement. The PbS QDs are thus promising contrast agents for fluorescence imaging and OCT

of cardiovascular disease.

Chapter 7. Conclusions and perspective

123

7.5. Future perspectives

During the doctoral thesis, the OCT setup has been mainly used for evaluating the potential

application of the NPs for cardiovascular imaging. The DFM is home built for visualization of the

NPs and cells, and in future, to see the internalized process of NPs in the cells.

In respect to the future work of contrast agent that have been used in my thesis, their basic

properties have been demonstrated. The future work would be the functionalization of these

contrast agents with special ligand so that they can be specially delivered to the cardiovascular

disease sites, and the evaluation of these contrast agents for ex and in vivo imaging of the

atherosclerosis process by OCT, fluorescence, and photoacoustic.

More contrast agent needs to be developed with higher scattering cross section than GNSs. In

addition, due to the strong scattering properties of the artery, absorption contrast agents are also

needed in order to create negative contrast in the artery, which could decrease the OCT signal and

thus indicate the cardiovascular disease sites.

Due to the catheter based IV-OCT system, the functionalized fluorescence NPs could be used

in the cardiovascular disease, as the catheter could not only introduce the light to the cardiovascular

disease and also collect the light from these areas. Thus the limitation of the fluorescence NPs such

as low penetration depth could be avoid. A combined OCT and fluorescence system needed to be

widely used for cardiovascular disease research.

Chapter 8. Conclusiones y

Perspectivas futuras (ES)

Chapter 8. Conclusiones y Perspectivas futuras (ES)

126

8.1. GNSs, mejor agente de contraste para OCT

Los diferentes tipos de GNPs aumenta de forma distinta el contraste de OCT debido a sus

diferentes propiedades de scattering, ya que el OCT se basa en el análisis de la señal de scattering.

Los GNSs con un tamaño alrededor de 220 nm muestran una sección eficaz de scattering mayor

que la de los GNRs y GNSTs a la longitud de onda del laser del OCT (1,3 µm), así que son los

mejores agentes de contraste para OCT entre todas las partículas estudiadas en esta tesis.

8.2. Visualización de GNSs individuales por IV-OCT

El IV-OCT puede suarse para seguir GNSs individuales gracias a la alta sensibilidad del IV-

OCT y a la alta sección eficaz de scattering de los GNSs.

8.3. Las GNSs pueden actuar como agentes de contraste para imagen celular y

aumento del contraste en tejidos por IV-OCT

El análisis del hostograma de intensidades de imágenes de OCT de GNSs incubadas en células

HeLa y Jurkat permite la detección de células vivas a las que se han adherido GNSs. El número de

células adheridas en función de la concentración de GNSs demostró una tendencia a la saturación

para dispersiones concentradas de GNSs. La internalización de las GNSs se ha confirmado por

TEM y DFM.

La inyección de las GNSs en un tejido muestra el aumento de ocntraste en la señal de OCT una

vez que se ha sustraído el fondo. Esto permite la visualización de la distribución de GNSs en el

tejido.

8.4. QDs de PbS pueden actuar como agentes de contraste duales para imagen

cardiovascular

La ancha banda de absorción de los QDs de PbS permiten la excitación de su luminiscencia con

el laser del OCT a 1.3 µm. Tras inyectar los QDs de PbS en la larteria, su emisión a 1,55 µm puede

usarse para obtener la imagen de fluorescencia en la tercera ventana biológica. Además, las

destacables propiedades de scattering de los QDs de PbS puede usarse para obtener contraste por

Chapter 8. Conclusiones y Perspectivas futuras (ES)

127

OCT. Los QDs de PbS son, por tanto, unos prometedores agentes de contraste para imagen de

fluorescencia y OCT en enfermedades cardiovasculares.

8.5. Perspectivas futuras

Durante la realización de esta tesis doctoral, la técnica de OCT se ha usado fundamentalmente

para evaluar el potencial usal de NPs como agentes de contraste para imagen cardiovascular. Se

ha desarrollado además un sistema de DFM para seguir el proceso de internalización de las NPs

en las células.

Respecto a las futuras aplicaciones de los agentes de contraste que se han usado en esta tesis,

sus propiedades fundamentales han sido demostradas. El trabajo futuro deberá estar encaminado a

la funcionalización de estos agentes de contraste con ligandos específicos, de forma que se dirijan

específicamente a regiones patológicas, y a la evaluación de dichos agentes de contraste en

imágenes in vivo y ex vivo de procesos de ateroesclerosis con medidas de OCT, fluorescencia y

fotoacústico.

Es necesario el desarrollo de más agentes de contraste con un scattering mayor que el de los

GNSs. Además, debido al fuerte scattering que se produce en las paredes de la arteria, sería

interesante el desarrollo de agentes e contraste basados en la absorción de luz, de forma que creen

un contraste negativo en la arteria, disminuyendo la señal de OCT e indicando la localización de

sitios patológicos en la arteria.

Debido al sistema de IV-OCT acoplado en un catéter, la fluorescencia de las NPs puede ser

recogida por el mismo catéter. De este modo, el problema de la baja penetración de la fluorescencia

puede resolverse. Un sistema combinado de OCT y fluorescencia debe ser usado para el estudio

de enfermedades cardiovasculares.

References

129

References

1. Development Organisation for Economic Co-Operation and. Health Status.

2018.http://stats.oecd.org/index.aspx?DataSetCode=HEALTH_STAT# (9 May 2018).

2. Libby Peter, Ridker Paul M and Hansson Göran K. Progress and challenges in translating

the biology of atherosclerosis. Nature. 2011;473:317-325.

3. Moore Kathryn J and Tabas Ira. Macrophages in the pathogenesis of atherosclerosis. Cell.

2011;145:341-355.

4. Zhang Jia, Zu Yujiao, Dhanasekara Chathurika S, Li Jun, Wu Dayong, Fan Zhaoyang and

Wang Shu. Detection and treatment of atherosclerosis using nanoparticles. Wiley Interdiscip Rev:

Nanomed Nanobiotechnol. 2017;9:e1412.

5. Tabas Ira, Williams Kevin Jon and Borén Jan. Subendothelial lipoprotein retention as the

initiating process in atherosclerosis. Circulation. 2007;116:1832-1844.

6. Sanz Javier and Fayad Zahi A. Imaging of atherosclerotic cardiovascular disease. Nature.

2008;451:953-957.

7. Mestas Javier and Ley Klaus. Monocyte-endothelial cell interactions in the development

of atherosclerosis. Trends in cardiovascular medicine. 2008;18:228-232.

8. Tabas Ira. Macrophage death and defective inflammation resolution in atherosclerosis. Nat

Rev Immunol. 2010;10:36-46.

9. Peiser Leanne, Mukhopadhyay Subhankar and Gordon Siamon. Scavenger receptors in

innate immunity. Curr Opin Immunol. 2002;14:123-128.

10. Järveläinen Hannu, Sainio Annele, Koulu Markku, Wight Thomas N and Penttinen Risto.

Extracellular matrix molecules: potential targets in pharmacotherapy. Pharmacological reviews.

2009;61:198-223.

11. Peng Wen-jia, Yan Jun-wei, Wan Ya-nan, Wang Bing-xiang, Tao Jin-hui, Yang Guo-jun,

Pan Hai-feng and Wang Jing. Matrix metalloproteinases: a review of their structure and role in

systemic sclerosis. Journal of clinical immunology. 2012;32:1409-1414.

12. Liu Jian, Sukhova Galina K, Sun Jiu-Song, Xu Wei-Hua, Libby Peter and Shi Guo-Ping.

Lysosomal cysteine proteases in atherosclerosis. Arterioscler, Thromb, Vasc Biol. 2004;24:1359-

1366.

13. Jones Charles B, Sane David C and Herrington David M. Matrix metalloproteinases: a

review of their structure and role in acute coronary syndrome. Cardiovasc Res. 2003;59:812-823.

14. Hansson Göran K. Inflammation, atherosclerosis, and coronary artery disease. N Engl J

Med. 2005;352:1685-1695.

15. Jaffer Farouc A and Weissleder Ralph. Molecular imaging in the clinical arena. Jama.

2005;293:855-862.

http://stats.oecd.org/index.aspx?DataSetCode=HEALTH_STAT

References

130

16. Wang Xiaowei and Peter Karlheinz. Molecular Imaging of Atherothrombotic

DiseasesHighlights. Arterioscler, Thromb, Vasc Biol. 2017;37:1029-1040.

17. Schlesinger Martin and Bendas Gerd. Vascular cell adhesion molecule‐1 (VCAM‐1)—

An increasing insight into its role in tumorigenicity and metastasis. Int J Cancer. 2015;136:2504-

2514.

18. Han Sung Gu, Newsome Bradley and Hennig Bernhard. Titanium dioxide nanoparticles

increase inflammatory responses in vascular endothelial cells. Toxicology. 2013;306:1-8.

19. Wen Song, Liu Dong-Fang, Cui Ying, Harris Steven Scott, Chen Yu-chen, Li King C, Ju

Sheng-hong and Teng Gao-Jun. In vivo MRI detection of carotid atherosclerotic lesions and

kidney inflammation in ApoE-deficient mice by using LOX-1 targeted iron nanoparticles.

Nanomedicine. 2014;10:639-649.

20. Jaffer Farouc A, Libby Peter and Weissleder Ralph. Molecular and cellular imaging of

atherosclerosis. J Am Coll Cardiol. 2006;47:1328-1338.

21. Huang D, Swanson EA, Lin CP, Schuman JS, Stinson WG, Chang W, Hee MR, Flotte T,

Gregory K and Puliafito CA. Optical coherence tomography. Science. 1991;254:1178-1181.

22. Drexler Wolfgang, Morgner Uwe, Ghanta Ravi K, Kärtner Franz X, Schuman Joel S and

Fujimoto James G. Ultrahigh-resolution ophthalmic optical coherence tomography. Nat Med.

2001;7:502-507.

23. Fujimoto James G. Optical coherence tomography for ultrahigh resolution in vivo imaging.

Nat Biotech. 2003;21:1361-1367.

24. Bezerra Hiram G, Costa Marco A, Guagliumi Giulio, Rollins Andrew M and Simon Daniel

I. Intracoronary optical coherence tomography: a comprehensive review: clinical and research

applications. JACC Cardiovasc Intv. 2009;2:1035-1046.

25. Jia Yali, Bailey Steven T, Hwang Thomas S, McClintic Scott M, Gao Simon S, Pennesi

Mark E, Flaxel Christina J, Lauer Andreas K, Wilson David J and Hornegger Joachim.

Quantitative optical coherence tomography angiography of vascular abnormalities in the living

human eye. Proc Natl Acad Sci. 2015;112:E2395-E2402.

26. Yabushita Hiroshi, Bouma Brett E, Houser Stuart L, Aretz H Thomas, Jang Ik-Kyung,

Schlendorf Kelly H, Kauffman Christopher R, Shishkov Milen, Kang Dong-Heon and Halpern

Elkan F. Characterization of human atherosclerosis by optical coherence tomography. Circulation.

2002;106:1640-1645.

27. Phipps Jennifer E, Hoyt Taylor, Milner Thomas E and Feldman Marc D. Translating

intravascular optical coherence tomography from a research to a clinical tool. Curr Cardiovasc

Imaging Rep. 2015;8:30.

28. Burke Allen P, Farb Andrew, Malcom Gray T, Liang You-hui, Smialek John and Virmani

Renu. Coronary risk factors and plaque morphology in men with coronary disease who died

suddenly. N Engl J Med. 1997;336:1276-1282.

29. Chia Stanley, Raffel Owen Christopher, Takano Masamichi, Tearney Guillermo J, Bouma

Brett E and Jang Ik-Kyung. Association of statin therapy with reduced coronary plaque rupture:

an optical coherence tomography study. Coronary artery disease. 2008;19:237-242.

References

131

30. Xu Chenyang, Schmitt Joseph M, Carlier Stephane G and Virmani Renu. Characterization

of atherosclerosis plaques by measuring both backscattering and attenuation coefficients in optical

coherence tomography. J Biomed Opt. 2008;13:034003.

31. Libby Peter, Ridker Paul M and Maseri Attilio. Inflammation and atherosclerosis.

Circulation. 2002;105:1135-1143.

32. Tearney Guillermo J, Yabushita Hiroshi, Houser Stuart L, Aretz H Thomas, Jang Ik-Kyung,

Schlendorf Kelly H, Kauffman Christopher R, Shishkov Milen, Halpern Elkan F and Bouma Brett

E. Quantification of macrophage content in atherosclerotic plaques by optical coherence

tomography. Circulation. 2003;107:113-119.

33. Di Vito Luca, Agozzino Manuela, Marco Valeria, Ricciardi Andrea, Concardi Monica,

Romagnoli Enrico, Gatto Laura, Calogero Giordano, Tavazzi Luigi and Arbustini Eloisa.

Identification and quantification of macrophage presence in coronary atherosclerotic plaques by

optical coherence tomography. European Heart Journal-Cardiovascular Imaging. 2015;16:807-

813.

34. Phipps Jennifer E, Vela Deborah, Hoyt Taylor, Halaney David L, Mancuso J Jacob, Buja

L Maximilian, Asmis Reto, Milner Thomas E and Feldman Marc D. Macrophages and

intravascular OCT bright spots: a quantitative study. JACC Cardiovasc Imaging. 2015;8:63-72.

35. Lowe Harry C, Narula Jagat, Fujimoto James G and Jang Ik-Kyung. Intracoronary optical

diagnostics: current status, limitations, and potential. JACC Cardiovasc Intv. 2011;4:1257-1270.

36. Bouma B. E., Tearney G. J., Yabushita H., Shishkov M., Kauffman C. R., DeJoseph

Gauthier D., MacNeill B. D., Houser S. L., Aretz H. T., Halpern E. F. and Jang I. K. Evaluation

of intracoronary stenting by intravascular optical coherence tomography. Heart. 2003;89:317-320.

37. Xu Minghua and Wang Lihong V. Photoacoustic imaging in biomedicine. Rev Sci Instrum.

2006;77:041101.

38. Jansen Krista, van Soest Gijs and van der Steen Antonius FW. Intravascular photoacoustic

imaging: a new tool for vulnerable plaque identification. Ultrasound in medicine and biology.

2014;40:1037-1048.

39. Luke Geoffrey P, Yeager Doug and Emelianov Stanislav Y. Biomedical applications of

photoacoustic imaging with exogenous contrast agents. Annals of biomedical engineering.

2012;40:422-437.

40. Laufer Jan, Delpy Dave, Elwell Clare and Beard Paul. Quantitative spatially resolved

measurement of tissue chromophore concentrations using photoacoustic spectroscopy: application

to the measurement of blood oxygenation and haemoglobin concentration. Phys Med Biol.

2007;52:141-168.

41. Laufer J, Zhang EZABPC and Beard P. Quantitative in-vivo measurements of blood

oxygen saturation using multiwavelength photoacoustic imaging. Photons Plus Ultrasound:

Imaging and Sensing 2007: The Eighth Conference on Biomedical Thermoacoustics,

Optoacoustics, and Acousto-optics. 2007;6437:64371Z.

42. Beard Paul. Biomedical photoacoustic imaging. Interface Focus. 2011:rsfs20110028.

43. Wang Lihong V. Prospects of photoacoustic tomography. Medical physics. 2008;35:5758-

5767.

References

132

44. Kruger Robert A, Lam Richard B, Reinecke Daniel R, Del Rio Stephen P and Doyle Ryan

P. Photoacoustic angiography of the breast. Medical physics. 2010;37:6096-6100.

45. Wang Bo, Su Jimmy L, Amirian James, Litovsky Silvio H, Smalling Richard and

Emelianov Stanislav. Detection of lipid in atherosclerotic vessels using ultrasound-guided

spectroscopic intravascular photoacoustic imaging. Opt Express. 2010;18:4889-4897.

46. Li Wanwan and Chen Xiaoyuan. Gold nanoparticles for photoacoustic imaging.

Nanomedicine. 2015;10:299-320.

47. Sangha Gurneet S and Goergen Craig J. Photoacoustic tomography: applications for

atherosclerosis imaging. Journal of Optics. 2016;18:084005.

48. Qin Huan, Zhao Yue, Zhang Jian, Pan Xiao, Yang Sihua and Xing Da. Inflammation-

targeted gold nanorods for intravascular photoacoustic imaging detection of matrix

metalloproteinase-2 (MMP2) in atherosclerotic plaques. Nanomedicine. 2016;12:1765-1774.

49. Hillman Elizabeth M. C., Amoozegar Cyrus B., Wang Tracy, McCaslin Addason F. H.,

Bouchard Matthew B., Mansfield James and Levenson Richard M. In vivo optical imaging and

dynamic contrast methods for biomedical research. Philosophical Transactions of the Royal

Society A: Mathematical, Physical and Engineering Sciences. 2011;369:4620-4643.

50. Ntziachristos Vasilis, Bremer Christoph and Weissleder Ralph. Fluorescence imaging with

near-infrared light: new technological advances that enable in vivo molecular imaging. European

radiology. 2003;13:195-208.

51. Jaque D, Maestro L Martinez, Del Rosal B, Haro-Gonzalez P, Benayas A, Plaza JL,

Rodriguez E Martin and Sole J Garcia. Nanoparticles for photothermal therapies. Nanoscale.

2014;6:9494-9530.

52. Hemmer Eva, Benayas Antonio, Légaré François and Vetrone Fiorenzo. Exploiting the

biological windows: current perspectives on fluorescent bioprobes emitting above 1000 nm.

Nanoscale Horiz. 2016;1:168-184.

53. Frangioni John V. In vivo near-infrared fluorescence imaging. Curr Opin Chem Biol.

2003;7:626-634.

54. Rao Jianghong, Dragulescu-Andrasi Anca and Yao Hequan. Fluorescence imaging in vivo:

recent advances. Curr Opin Biotechnol. 2007;18:17-25.

55. Hong Guosong, Antaris Alexander L and Dai Hongjie. Near-infrared fluorophores for

biomedical imaging. Nat Biomed Eng. 2017;1:0010.

56. Jaffer Farouc A, Vinegoni Claudio, John Michael C, Aikawa Elena, Gold Herman K, Finn

Aloke V, Ntziachristos Vasilis, Libby Peter and Weissleder Ralph. Real-time catheter molecular

sensing of inflammation in proteolytically active atherosclerosis. Circulation. 2008;118:1802-

1809.

57. Jaffer Farouc A, Calfon Marcella A, Rosenthal Amir, Mallas Georgios, Razansky R Nika,

Mauskapf Adam, Weissleder Ralph, Libby Peter and Ntziachristos Vasilis. Two-dimensional

intravascular near-infrared fluorescence molecular imaging of inflammation in atherosclerosis and

stent-induced vascular injury. J Am Coll Cardiol. 2011;57:2516-2526.

References

133

58. Dweck Marc R, Doris Mhairi K, Motwani Manish, Adamson Philip D, Slomka Piotr, Dey

Damini, Fayad Zahi A, Newby David E and Berman Daniel. Imaging of coronary

atherosclerosis—evolution towards new treatment strategies. Nature Reviews Cardiology.

2016;13:533-548.

59. Vinegoni Claudio, Botnaru Ion, Aikawa Elena, Calfon Marcella A, Iwamoto Yoshiko,

Folco Eduardo J, Ntziachristos Vasilis, Weissleder Ralph, Libby Peter and Jaffer Farouc A.

Indocyanine green enables near-infrared fluorescence imaging of lipid-rich, inflamed

atherosclerotic plaques. Science translational medicine. 2011;3:84ra45.

60. Yoo Hongki, Kim Jin Won, Shishkov Milen, Namati Eman, Morse Theodore, Shubochkin

Roman, McCarthy Jason R, Ntziachristos Vasilis, Bouma Brett E and Jaffer Farouc A. Intra-

arterial catheter for simultaneous microstructural and molecular imaging in vivo. Nat Med.

2011;17:1680-1684.

61. Padmanabhan Parasuraman, Kumar Ajay, Kumar Sundramurthy, Chaudhary Ravi Kumar

and Gulyás Balázs. Nanoparticles in practice for molecular-imaging applications: An overview.

Acta Biomaterialia. 2016;41:1-16.

62. Brigger Irene, Dubernet Catherine and Couvreur Patrick. Nanoparticles in cancer therapy

and diagnosis. Adv Drug Delivery Rev. 2012;64:24-36.

63. Saraste Antti, Nekolla Stephan G and Schwaiger Markus. Cardiovascular molecular

imaging: an overview. Cardiovasc Res. 2009;83:643-652.

64. Chen Ian Y and Wu Joseph C. Cardiovascular molecular imaging: focus on clinical

translation. Circulation. 2011;123:425-443.

65. Sinusas Albert J, Bengel Frank, Nahrendorf Matthias, Epstein Frederick H, Wu Joseph C,

Villanueva Flordeliza S, Fayad Zahi A and Gropler Robert J. Multimodality cardiovascular

molecular imaging, part I. Circ Cardiovasc Imaging. 2008;1:244-256.

66. Nguyen Luong TH, Muktabar Aristo, Tang Jinkai, Dravid Vinayak P, Thaxton C Shad,

Venkatraman Subbu and Ng Kee Woei. Engineered nanoparticles for the detection, treatment and

prevention of atherosclerosis: how close are we? Drug Discovery Today. 2017;22:1438-1446.

67. Mulder Willem JM, Jaffer Farouc A, Fayad Zahi A and Nahrendorf Matthias. Imaging and

nanomedicine in inflammatory atherosclerosis. Science translational medicine. 2014;6:239sr1.

68. Cicha Iwona, Lyer Stefan, Alexiou Christoph and Garlichs Christoph D. Nanomedicine in

diagnostics and therapy of cardiovascular diseases: beyond atherosclerotic plaque imaging.

Nanotechnol Rev. 2013;2:449-472.

69. Varna Mariana, Xuan Hoa V and Fort Emmanuel. Gold nanoparticles in cardiovascular

imaging. Wiley Interdiscip Rev: Nanomed Nanobiotechnol. 2018;10:e1470.

70. Liz-Marzán Luis M. Nanometals: formation and color. Mater Today. 2004;7:26-31.

71. Link Stephan, Mohamed MB and El-Sayed MA. Simulation of the optical absorption

spectra of gold nanorods as a function of their aspect ratio and the effect of the medium dielectric

constant. The Journal of Physical Chemistry B. 1999;103:3073-3077.

72. Mayer Kathryn M and Hafner Jason H. Localized surface plasmon resonance sensors.

Chem Rev. 2011;111:3828-3857.

References

134

73. Mie Gustav. Beiträge zur Optik trüber Medien, speziell kolloidaler Metallösungen.

Annalen der physik. 1908;330:377-445.

74. Gans R. Über die form ultramikroskopischer silberteilchen. Annalen der Physik.

1915;352:270-284.

75. Zhao Jing, Pinchuk Anatoliy O, McMahon Jeffrey M, Li Shuzhou, Ausman Logan K,

Atkinson Ariel L and Schatz George C. Methods for describing the electromagnetic properties of

silver and gold nanoparticles. Acc Chem Res. 2008;41:1710-1720.

76. Draine Bruce T and Flatau Piotr J. Discrete-dipole approximation for scattering

calculations. JOSA A. 1994;11:1491-1499.

77. Jain Prashant K, Lee Kyeong Seok, El-Sayed Ivan H and El-Sayed Mostafa A. Calculated

absorption and scattering properties of gold nanoparticles of different size, shape, and composition:

applications in biological imaging and biomedicine. The journal of physical chemistry B.

2006;110:7238-7248.

78. Dykman Lev and Khlebtsov Nikolai. Gold nanoparticles in biomedical applications: recent

advances and perspectives. Chem Soc Rev. 2012;41:2256-2282.

79. Huang Xiaohua, Neretina Svetlana and El‐Sayed Mostafa A. Gold nanorods: from

synthesis and properties to biological and biomedical applications. Adv Mater. 2009;21:4880-4910.

80. Amendola Vincenzo, Pilot Roberto, Frasconi Marco, Maragò Onofrio M and Iatì Maria

Antonia. Surface plasmon resonance in gold nanoparticles: a review. J Phys: Condens Matter.

2017;29:203002.

81. Zerda Adam, Prabhulkar Shradha, Perez Victor L, Ruggeri Marco, Paranjape Amit S,

Habte Frezghi, Gambhir Sanjiv S and Awdeh Richard M. Optical coherence contrast imaging

using gold nanorods in living mice eyes. Clin Exp Ophthalmol. 2015;43:358-366.

82. Chen Jingyi, Saeki Fusayo, Wiley Benjamin J, Cang Hu, Cobb Michael J, Li Zhi-Yuan,

Au Leslie, Zhang Hui, Kimmey Michael B and Li Xingde. Gold nanocages: bioconjugation and

their potential use as optical imaging contrast agents. Nano Lett. 2005;5:473-477.

83. Gao Xiaohu, Cui Yuanyuan, Levenson Richard M, Chung Leland WK and Nie Shuming.

In vivo cancer targeting and imaging with semiconductor quantum dots. Nat Biotech. 2004;22:969.

84. Michalet X, Pinaud FF, Bentolila LA, Tsay JM, Doose SJJL, Li JJ, Sundaresan G, Wu AM,

Gambhir SS and Weiss S. Quantum dots for live cells, in vivo imaging, and diagnostics. Science.

2005;307:538-544.

85. Valizadeh Alireza, Mikaeili Haleh, Samiei Mohammad, Farkhani Samad Mussa, Zarghami

Nosratalah, Akbarzadeh Abolfazl and Davaran Soodabeh. Quantum dots: synthesis,

bioapplications, and toxicity. Nanoscale Res Lett. 2012;7:480.

86. Wolfbeis Otto S. An overview of nanoparticles commonly used in fluorescent bioimaging.

Chem Soc Rev. 2015;44:4743-4768.

87. Burda Clemens, Chen Xiaobo, Narayanan Radha and El-Sayed Mostafa A. Chemistry and

properties of nanocrystals of different shapes. Chem Rev. 2005;105:1025-1102.

88. Koole Rolf, Groeneveld Esther, Vanmaekelbergh Daniel, Meijerink Andries and de Mello

Donegá Celso. Size effects on semiconductor nanoparticles Nanoparticles: Springer; 2014: 13-51.

References

135

89. Efros Al L and Rosen M. The electronic structure of semiconductor nanocrystals. Annu

Rev Mater Sci. 2000;30:475-521.

90. Chukwuocha Ephrem O, Onyeaju Michael C and Harry Taylor ST. Theoretical studies on

the effect of confinement on quantum dots using the brus equation. World Journal of Condensed

Matter Physics. 2012;2:96-100.

91. Krauss Todd D. and Peterson Jeffrey J. Electronic structure and optical transitions in

colloidal semiconductor nanocrystals. In: E. H. Sargent and G. Konstantatos, eds. Colloidal

Quantum Dot Optoelectronics and Photovoltaics Cambridge: Cambridge University Press; 2013:

59-86.

92. Madelung Otfried. Semiconductors: data handbook. 3rd ed. New York: Springer Science

& Business Media; 2012.

93. Algar W. Russ, Susumu Kimihiro, Delehanty James B. and Medintz Igor L. Semiconductor

Quantum Dots in Bioanalysis: Crossing the Valley of Death. Anal Chem. 2011;83:8826-8837.

94. Cheng Cheng, Li Jiejie and Cheng Xiaoyu. Photoluminescence lifetime and absorption

spectrum of PbS nanocrystal quantum dots. J Lumin. 2017;188:252-257.

95. Wegner K David and Hildebrandt Niko. Quantum dots: bright and versatile in vitro and in

vivo fluorescence imaging biosensors. Chem Soc Rev. 2015;44:4792-4834.

96. Zhou Juan, Yang Yong and Zhang Chun-yang. Toward biocompatible semiconductor

quantum dots: from biosynthesis and bioconjugation to biomedical application. Chem Rev.

2015;115:11669-11717.

97. Kobayashi Hisataka, Hama Yukihiro, Koyama Yoshinori, Barrett Tristan, Regino Celeste

AS, Urano Yasuteru and Choyke Peter L. Simultaneous multicolor imaging of five different

lymphatic basins using quantum dots. Nano Lett. 2007;7:1711-1716.

98. Wu Chenxin, Zhang Yejun, Li Zhen, Li Chunyan and Wang Qiangbin. A novel

photoacoustic nanoprobe of ICG@ PEG-Ag2S for atherosclerosis targeting and imaging in vivo.

Nanoscale. 2016;8:12531-12539.

99. Hong Guosong, Robinson Joshua T, Zhang Yejun, Diao Shuo, Antaris Alexander L, Wang

Qiangbin and Dai Hongjie. In Vivo Fluorescence Imaging with Ag2S Quantum Dots in the Second

Near‐Infrared Region. Angew Chem. 2012;124:9956-9959.

100. Kelley William J, Safari Hanieh, Lopez‐Cazares Genesis and Eniola‐Adefeso Omolola.

Vascular‐targeted nanocarriers: design considerations and strategies for successful treatment of

atherosclerosis and other vascular diseases. Wiley Interdiscip Rev: Nanomed Nanobiotechnol.

2016;8:909-926.

101. Blanco Elvin, Shen Haifa and Ferrari Mauro. Principles of nanoparticle design for

overcoming biological barriers to drug delivery. Nat Biotech. 2015;33:941-951.

102. Nel Andre E, Mädler Lutz, Velegol Darrell, Xia Tian, Hoek Eric MV, Somasundaran

Ponisseril, Klaessig Fred, Castranova Vince and Thompson Mike. Understanding

biophysicochemical interactions at the nano–bio interface. Nat Mater. 2009;8:543-557.

References

136

103. Kanwar Rupinder K, Chaudhary Rajneesh, Tsuzuki Takuya and Kanwar Jagat R. Emerging

engineered magnetic nanoparticulate probes for targeted MRI of atherosclerotic plaque

macrophages. Nanomedicine. 2012;7:735-749.

104. Kooi ME, Cappendijk VC, Cleutjens KBJM, Kessels AGH, Kitslaar PJEHM, Borgers M,

Frederik PM, Daemen MJAP and Van Engelshoven JMA. Accumulation of ultrasmall

superparamagnetic particles of iron oxide in human atherosclerotic plaques can be detected by in

vivo magnetic resonance imaging. Circulation. 2003;107:2453-2458.

105. Qiao Ruirui, Qiao Hongyu, Zhang Yan, Wang Yabin, Chi Chongwei, Tian Jie, Zhang

Lifang, Cao Feng and Gao Mingyuan. Molecular imaging of vulnerable atherosclerotic plaques in

vivo with osteopontin-specific upconversion nanoprobes. ACS Nano. 2017;11:1816-1825.

106. Jefferson Andrew, Wijesurendra Rohan S, McAteer Martina A, Digby Janet E, Douglas

Gillian, Bannister Thomas, Perez-Balderas Francisco, Bagi Zsolt, Lindsay Alistair C and

Choudhury Robin P. Molecular imaging with optical coherence tomography using ligand-

conjugated microparticles that detect activated endothelial cells: rational design through target

quantification. Atherosclerosis. 2011;219:579-587.

107. Jaffer Farouc A., Tung Ching-Hsuan, Gerszten Robert E. and Weissleder Ralph. In Vivo

Imaging of Thrombin Activity in Experimental Thrombi With Thrombin-Sensitive Near-Infrared

Molecular Probe. Arterioscler, Thromb, Vasc Biol. 2002;22:1929-1935.

108. Peters David, Kastantin Mark, Kotamraju Venkata Ramana, Karmali Priya P, Gujraty

Kunal, Tirrell Matthew and Ruoslahti Erkki. Targeting atherosclerosis by using modular,

multifunctional micelles. Proc Natl Acad Sci. 2009;106:9815-9819.

109. Wang Yabin, Chen Jiangwei, Yang Bo, Qiao Hongyu, Gao Lei, Su Tao, Ma Sai, Zhang

Xiaotian, Li Xiujuan and Liu Gang. In vivo MR and fluorescence dual-modality imaging of

atherosclerosis characteristics in mice using profilin-1 targeted magnetic nanoparticles.

Theranostics. 2016;6:272-286.

110. Wu Chenxin, Zhang Yejun, Li Zhen, Li Chunyan and Wang Qiangbin. A novel

photoacoustic nanoprobe of ICG@ PEG-Ag 2 S for atherosclerosis targeting and imaging in vivo.

Nanoscale. 2016;8:12531-12539.

111. Qin Huan, Zhou Ting, Yang Sihua, Chen Qun and Xing Da. Gadolinium (III)-gold

nanorods for MRI and photoacoustic imaging dual-modality detection of macrophages in

atherosclerotic inflammation. Nanomedicine. 2013;8:1611-1624.

112. Kim Kang, Huang Sheng-Wen, Ashkenazi Shai, O’Donnell Matthew, Agarwal Ashish,

Kotov Nicholas A, Denny Michael F and Kaplan Mariana J. Photoacoustic imaging of early

inflammatory response using gold nanorods. Appl Phys Lett. 2007;90:223901.

113. Rouleau Leonie, Berti Romain, Ng Vanessa WK, Matteau‐Pelletier Carl, Lam Tina,

Saboural Pierre, Kakkar Ashok K, Lesage Frédéric, Rhéaume Eric and Tardif Jean‐Claude.

VCAM‐1‐targeting gold nanoshell probe for photoacoustic imaging of atherosclerotic plaque

in mice. Contrast Media Mol Imaging. 2013;8:27-39.

114. Wang Bo, Yantsen Evgeniya, Larson Timothy, Karpiouk Andrei B, Sethuraman Shriram,

Su Jimmy L, Sokolov Konstantin and Emelianov Stanislav Y. Plasmonic intravascular

References

137

photoacoustic imaging for detection of macrophages in atherosclerotic plaques. Nano Lett.

2009;9:2212-2217.

115. Ha Seunghan, Carson Andrew, Agarwal Ashish, Kotov Nicholas A and Kim Kang.

Detection and monitoring of the multiple inflammatory responses by photoacoustic molecular

imaging using selectively targeted gold nanorods. Biomed Opt Express. 2011;2:645-657.

116. Hu Jie, Ortgies Dirk H., Torres Rio Aguliar, Fernández Nuria, Porto Lucas, Rodríguez

Emma Martín, Solé José García, Jaque Daniel, Alfonso Fernando and Rivero Fernando. Quantum

Dots Emitting in the Third Biological Window as Bimodal Contrast Agents for Cardiovascular

Imaging. Adv Funct Mater. 2017;27:1703276.

117. Paranjape Amit S, Kuranov Roman, Baranov Stepan, Ma Li Leo, Villard Joseph W, Wang

Tianyi, Sokolov Konstantin V, Feldman Marc D, Johnston Keith P and Milner Thomas E. Depth

resolved photothermal OCT detection of macrophages in tissue using nanorose. Biomed Opt

Express. 2010;1:2-16.

118. Kim Jongsik, Ahmad Adeel, Li Joanne, Marjanovic Marina, Chaney Eric J, Suslick

Kenneth S and Boppart Stephen A. Intravascular magnetomotive optical coherence tomography

of targeted early‐ stage atherosclerotic changes in ex vivo hyperlipidemic rabbit aortas. J

Biophotonics. 2016;9:109-116.

119. Morishige Kunio, Kacher Daniel F, Libby Peter, Josephson Lee, Ganz Peter, Weissleder

Ralph and Aikawa Masanori. High-Resolution Magnetic Resonance Imaging Enhanced With

Superparamagnetic Nanoparticles Measures Macrophage Burden in AtherosclerosisClinical

Perspective. Circulation. 2010;122:1707-1715.

120. Marrache Sean and Dhar Shanta. Biodegradable synthetic high-density lipoprotein

nanoparticles for atherosclerosis. Proc Natl Acad Sci. 2013;110:9445-9450.

121. Yuan Hsiangkuo, Khoury Christopher G, Hwang Hanjun, Wilson Christy M, Grant Gerald

A and Vo-Dinh Tuan. Gold nanostars: surfactant-free synthesis, 3D modelling, and two-photon

photoluminescence imaging. Nanotechnology. 2012;23:075102.

122. Williams David B and Carter C Barry. The transmission electron microscope Transmission

electron microscopy: Springer; 1996: 3-17.

123. Kregsamer Peter, Streli Christina and Wobrauschek Peter. Total reflection X-ray

fluorescence. PRACTICAL SPECTROSCOPY SERIES. 2002;29:559-602.

124. Klockenkàmper Rinhold, Knoth Joachim, Prange Andreas and Schwenke Heinrich. Total-

reflection X-ray fluorescence spectroscopy. Anal Chem. 1992;64:1115A-1123A.

125. del Rosal Blanca, Pérez ‐ Delgado Alberto, Carrasco Elisa, Jovanović Dragana J,

Dramićanin Miroslav D, Dražić Goran, de la Fuente Ángeles Juarranz, Sanz ‐ Rodriguez

Francisco and Jaque Daniel. Neodymium‐Based Stoichiometric Ultrasmall Nanoparticles for

Multifunctional Deep-Tissue Photothermal Therapy. Adv Opt Mater. 2016;4:782-789.

126. Roper D Keith, Ahn Wonmi and Hoepfner M. Microscale heat transfer transduced by

surface plasmon resonant gold nanoparticles. J Phys Chem C. 2007;111:3636-3641.

References

138

127. Izatt Joseph A, Choma Michael A and Dhalla Al-Hafeez. Theory of optical coherence

tomography. In: W. Drexler and J. G. Fujimoto, eds. Optical Coherence Tomography: Technology

and Applications: Springer, Cham; 2015: 65-94.

128. Jang Ik-Kyung, Bouma Brett E, Kang Dong-Heon, Park Seung-Jung, Park Seong-Wook,

Seung Ki-Bae, Choi Kyu-Bo, Shishkov Milen, Schlendorf Kelly and Pomerantsev Eugene.

Visualization of coronary atherosclerotic plaques in patients using optical coherence tomography:

comparison with intravascular ultrasound. J Am Coll Cardiol. 2002;39:604-609.

129. Leggio Luca, Gallego Daniel, Arroyo Raul, Gawali Sandeep Babu, Rodriguez Sergio,

Sánchez Miguel, Carpintero del Barrio Guillermo and Lamela Horacio. Accurate Determination

of Gold Nanorods Concentrations from Optoacoustic Signals Detected at 870 nm and 905 nm by

Using High-Power Diode Lasers with Fast Switching Electronics. Prog Electromagn Res.

2017;78:143-154.

130. Masters John R. HeLa cells 50 years on: the good, the bad and the ugly. Nat Rev Cancer.

2002;2:315-319.

131. Riss Terry L, Moravec Richard A, Niles Andrew L, Duellman Sarah, Benink Hélène A,

Worzella Tracy J and Minor Lisa. Cell viability assays Assay Guidance Manual Bethesda: Eli Lilly

& Company and the National Center for Advancing Translational Sciences; 2016: 355-386.

132. Ehlers Justis P, Gupta Preeya K, Farsiu Sina, Maldonado Ramiro, Kim Terry, Toth Cynthia

A and Mruthyunjaya Prithvi. Evaluation of contrast agents for enhanced visualization in optical

coherence tomography. Invest Ophthalmol Visual Sci. 2010;51:6614-6619.

133. Barton Jennifer Kehlet, Hoying James B and Sullivan Chris J. Use of microbubbles as an

optical coherence tomography contrast agent. Acad Radiol. 2002;9:S52-S55.

134. Dreaden Erik C, Alkilany Alaaldin M, Huang Xiaohua, Murphy Catherine J and El-Sayed

Mostafa A. The golden age: gold nanoparticles for biomedicine. Chem Soc Rev. 2012;41:2740-

2779.

135. Pérez-Juste Jorge, Pastoriza-Santos Isabel, Liz-Marzán Luis M and Mulvaney Paul. Gold

nanorods: synthesis, characterization and applications. Coord Chem Rev. 2005;249:1870-1901.

136. Van Hoecke Karen, De Schamphelaere KAC, Ali Zulqurnain, Zhang Feng, Elsaesser

Andreas, Rivera-Gil Pilar, Parak WJ, Smagghe Guy, Howard CV and Janssen CR. Ecotoxicity and

uptake of polymer coated gold nanoparticles. Nanotoxicology. 2013;7:37-47.

137. Rivera_Gil Pilar, Vazquez-Vazquez Carmen, Giannini Vincenzo, Callao M Pilar, Parak

Wolfgang J, Correa-Duarte Miguel A and Alvarez-Puebla Ramon A. Plasmonic Nanoprobes for

Real‐Time Optical Monitoring of Nitric Oxide inside Living Cells. Angew Chem Int Ed.

2013;52:13694-13698.

138. Zamora-Perez Paula, Tsoutsi Dionysia, Xu Ruixue and Rivera_Gil Pilar. Hyperspectral-

Enhanced Dark Field Microscopy for Single and Collective Nanoparticle Characterization in

Biological Environments. Materials. 2018;11:243.

139. Dionysia Tsoutsi, Marcos Sanles-Sobrido, Andreu Cabo and Pilar Rivera Gil. Common

aspects influencing the translocation of SERS to Biomedicine. Curr Med Chem. 2018;25:1-15.

140. Hainfeld JF, Slatkin DN, Focella TM and Smilowitz HM. Gold nanoparticles: a new X-ray

contrast agent. Br J Radiol. 2006;79:248-253.

References

139

141. Kim Dongkyu, Park Sangjin, Lee Jae Hyuk, Jeong Yong Yeon and Jon Sangyong.

Antibiofouling polymer-coated gold nanoparticles as a contrast agent for in vivo X-ray computed

tomography imaging. J Am Chem Soc. 2007;129:7661-7665.

142. Chhour Peter, Naha Pratap C, O'Neill Sean M, Litt Harold I, Reilly Muredach P, Ferrari

Victor A and Cormode David P. Labeling monocytes with gold nanoparticles to track their

recruitment in atherosclerosis with computed tomography. Biomaterials. 2016;87:93-103.

143. Oldenburg Amy L, Hansen Matthew N, Zweifel Daniel A, Wei Alexander and Boppart

Stephen A. Plasmon-resonant gold nanorods as low backscattering albedo contrast agents. Opt

Express. 2006;14:6724-6738.

144. Muskens Otto L, Bachelier Guillaume, Fatti Natalia Del, Vallee Fabrice, Brioude Arnaud,

Jiang Xuchuan and Pileni Marie-Paule. Quantitative absorption spectroscopy of a single gold

nanorod. J Phys Chem C. 2008;112:8917-8921.

145. Gudjonson Herman, Kats Mikhail A, Liu Kun, Nie Zhihong, Kumacheva Eugenia and

Capasso Federico. Accounting for inhomogeneous broadening in nano-optics by electromagnetic

modeling based on Monte Carlo methods. Proc Natl Acad Sci. 2014;111:E639-E644.

146. Oldenburg Steven J, Jackson Joseph B, Westcott Sarah L and Halas NJ. Infrared extinction

properties of gold nanoshells. Appl Phys Lett. 1999;75:2897-2899.

147. Nehl Colleen L, Grady Nathaniel K, Goodrich Glenn P, Tam Felicia, Halas Naomi J and

Hafner Jason H. Scattering spectra of single gold nanoshells. Nano Lett. 2004;4:2355-2359.

148. Agrawal Anant, Huang Stanley, Lin Alex Wei Han, Lee Min-Ho, Barton Jennifer Kehlet,

Drezek Rebekah Anna and Pfefer Joshua. Quantitative evaluation of optical coherence

tomography signal enhancement with gold nanoshells. J Biomed Opt. 2006;11:041121.

149. Gobin André M, Lee Min Ho, Halas Naomi J, James William D, Drezek Rebekah A and

West Jennifer L. Near-infrared resonant nanoshells for combined optical imaging and

photothermal cancer therapy. Nano Lett. 2007;7:1929-1934.

150. Maestro Laura Martínez, Haro-González Patricia, Coello Jose Garcia and Jaque Daniel.

Absorption efficiency of gold nanorods determined by quantum dot fluorescence thermometry.

Appl Phys Lett. 2012;100:201110.

151. Kah James Chen-Yong, Olivo Malini C, Chow Tzu Hao, San Song Kin, Koh Karen Zhen

Yu, Mhaisalkar Subodh and Sheppard Colin. Control of optical contrast using gold nanoshells for

optical coherence tomography imaging of mouse xenograft tumor model in vivo. J Biomed Opt.

2009;14:054015.

152. Yi Ji, Radosevich Andrew J, Rogers Jeremy D, Norris Sam CP, Çapoğlu İlker R, Taflove

Allen and Backman Vadim. Can OCT be sensitive to nanoscale structural alterations in biological

tissue? Opt Express. 2013;21:9043-9059.

153. Alexandrov Sergey, Subhash Hrebesh and Leahy Martin. Nanosensitive optical coherence

tomography for the study of changes in static and dynamic structures. Quantum Electron.

2014;44:657.

154. Alexandrov Sergey A, Subhash Hrebesh M, Zam Azhar and Leahy Martin. Nano-sensitive

optical coherence tomography. Nanoscale. 2014;6:3545-3549.

References

140

155. Liu Yang, Ashton Jeffrey R, Moding Everett J, Yuan Hsiangkuo, Register Janna K, Fales

Andrew M, Choi Jaeyeon, Whitley Melodi J, Zhao Xiaoguang and Qi Yi. A plasmonic gold

nanostar theranostic probe for in vivo tumor imaging and photothermal therapy. Theranostics.

2015;5:946-960.

156. Cheng Fong-Yu, Chen Chen-Tai and Yeh Chen-Sheng. Comparative efficiencies of

photothermal destruction of malignant cells using antibody-coated silica@ Au nanoshells, hollow

Au/Ag nanospheres and Au nanorods. Nanotechnology. 2009;20:425104.

157. Chen Huanjun, Shao Lei, Ming Tian, Sun Zhenhua, Zhao Chunmei, Yang Baocheng and

Wang Jianfang. Understanding the photothermal conversion efficiency of gold nanocrystals. Small.

2010;6:2272-2280.

158. Pattani Varun P and Tunnell James W. Nanoparticle‐mediated photothermal therapy: A

comparative study of heating for different particle types. Lasers Surg Med. 2012;44:675-684.

159. Wojtkowski Maciej, Kowalczyk A, Leitgeb R and Fercher AF. Full range complex spectral

optical coherence tomography technique in eye imaging. Opt Lett. 2002;27:1415-1417.

160. Drexler Wolfgang and Fujimoto James G. State-of-the-art retinal optical coherence

tomography. Prog Retinal Eye Res. 2008;27:45-88.

161. Cole Joseph R, Mirin Nikolay A, Knight Mark W, Goodrich Glenn P and Halas Naomi J.

Photothermal efficiencies of nanoshells and nanorods for clinical therapeutic applications. J Phys

Chem C. 2009;113:12090-12094.

162. Rodríguez-Oliveros R and Sánchez-Gil José A. Gold nanostars as thermoplasmonic

nanoparticles for optical heating. Opt Express. 2012;20:621-626.

163. Rodríguez-Oliveros Rogelio, Paniagua-Domínguez Ramón, Sánchez-Gil José A and

Macías Demetrio. Plasmon spectroscopy: Theoretical and numerical calculations, and

optimization techniques. Nanospectroscopy. 2014;1:67-96.

164. Marciniak Lukasz, Pilch Aleksandra, Arabasz Sebastian, Jin Dayong and Bednarkiewicz

Artur. Heterogeneously Nd3+ doped single nanoparticles for NIR-induced heat conversion,

luminescence, and thermometry. Nanoscale. 2017;9:8288-8297.

165. Hu Min, Novo Carolina, Funston Alison, Wang Haining, Staleva Hristina, Zou Shengli,

Mulvaney Paul, Xia Younan and Hartland Gregory V. Dark-field microscopy studies of single

metal nanoparticles: understanding the factors that influence the linewidth of the localized surface

plasmon resonance. J Mater Chem. 2008;18:1949-1960.

166. Reid MT Homer and Johnson Steven G. Efficient computation of power, force, and torque

in BEM scattering calculations. IEEE Trans Antennas Propag. 2015;63:3588-3598.

167. Ungureanu C, Gopal R Raja, van Leeuwen TG and Manohar Srirang. Optical properties of

gold nanorods: towards contrast agents for Photoacoustic imaging. Phys Rev B. 1972;6:4370-4379.

168. Prodan E, Nordlander P and Halas NJ. Electronic structure and optical properties of gold

nanoshells. Nano Lett. 2003;3:1411-1415.

169. Westcott S. L., Jackson J. B., Radloff C. and Halas N. J. Relative contributions to the

plasmon line shape of metal nanoshells. Phys Rev B. 2002;66:155431.

References

141

170. Cunningham Vincent B, Lamela Horacio and Gallego Daniel C. Laser optoacoustic scheme

for highly accurate characterization of gold nanostructures in liquid phantoms for biomedical

applications. J Nanophotonics. 2013;7:073078.

171. Cox Benjamin T, Laufer Jan G, Beard Paul C and Arridge Simon R. Quantitative

spectroscopic photoacoustic imaging: a review. J Biomed Opt. 2012;17:061202.

172. Kennedy Brendan F., Kennedy Kelsey M., Oldenburg Amy L., Adie Steven G., Boppart

Stephen A. and Sampson David D. Optical Coherence Elastography. In: W. Drexler and J. G.

Fujimoto, eds. Optical Coherence Tomography: Technology and Applications Cham: Springer

International Publishing; 2015: 1007-1054.

173. Alfonso Fernando, Dutary Jaime, Paulo Manuel, Gonzalo Nieves, Pérez-Vizcayno Maria

J, Jiménez-Quevedo Pilar, Escaned Javier, Bañuelos Camino, Hernández Rosana and Macaya

Carlos. Combined use of optical coherence tomography and intravascular ultrasound imaging in

patients undergoing coronary interventions for stent thrombosis. Heart. 2012;98:1213-1220.

174. Prati Francesco, Stazi Filippo, Dutary Jaime, La Manna Alessio, Di Giorgio Alessandro,

Pawlosky Tomasz, Gonzalo Nieves, Di Salvo Maria Elena, Imola Fabrizio and Tamburino Corrado.

Detection of very early stent healing after primary angioplasty: an optical coherence tomographic

observational study of chromium cobaltum and first-generation drug-eluting stents. The

DETECTIVE study. Heart. 2011;97:1841-1846.

175. Rivero Fernando, Bastante Teresa, Cuesta Javier, Benedicto Amparo, Restrepo Jorge A

and Alfonso Fernando. Treatment of in-stent restenosis with bioresorbable vascular scaffolds:

optical coherence tomography insights. Can J Cardiol. 2015;31:255-259.

176. Douma Kim, Prinzen Lenneke, Slaaf Dick W, Reutelingsperger Chris PM, Biessen Erik

AL, Hackeng Tilman M, Post Mark J and van Zandvoort Marc AMJ. Nanoparticles for optical

molecular imaging of atherosclerosis. Small. 2009;5:544-557.

177. Bibikova Olga, Popov Alexey, Bykov Alexander, Prilepskii Artur, Kinnunen Matti,

Kordas Krisztian, Bogatyrev Vladimir, Khlebtsov Nikolai, Vainio Seppo and Tuchin Valery.

Optical properties of plasmon-resonant bare and silica-coated nanostars used for cell imaging. J

Biomed Opt. 2015;20:076017.

178. Skrabalak Sara E, Chen Jingyi, Au Leslie, Lu Xianmao, Li Xingde and Xia Younan. Gold

nanocages for biomedical applications. Adv Mater. 2007;19:3177-3184.

179. Hu Jie, Rivero Fernando, Torres Rio Aguilar, Loro Ramírez Héctor, Rodríguez Emma

Martín, Alfonso Fernando, García Solé José and Jaque Daniel. Dynamic single gold nanoparticle

visualization by clinical intracoronary optical coherence tomography. J Biophotonics.

2017;10:674-682.

180. Skala Melissa C, Crow Matthew J, Wax Adam and Izatt Joseph A. Photothermal optical

coherence tomography of epidermal growth factor receptor in live cells using immunotargeted

gold nanospheres. Nano Lett. 2008;8:3461-3467.

181. Loo Christopher, Lowery Amanda, Halas Naomi, West Jennifer and Drezek Rebekah.

Immunotargeted nanoshells for integrated cancer imaging and therapy. Nano Lett. 2005;5:709-711.

References

142

182. De León Y Ponce, Pichardo-Molina JL, Ochoa N Alcalá and Luna-Moreno D. Contrast

enhancement of optical coherence tomography images using branched gold nanoparticles. J

Nanomater. 2012;2012:128.

183. Adler Desmond C, Huang Shu-Wei, Huber Robert and Fujimoto James G. Photothermal

detection of gold nanoparticles using phase-sensitive optical coherence tomography. Opt Express.

2008;16:4376-4393.

184. Fratoddi Ilaria, Venditti Iole, Cametti Cesare and Russo Maria Vittoria. How toxic are gold

nanoparticles? The state-of-the-art. Nano Res. 2015;8:1771-1799.

185. Meng X., Yang J., Dong M., Zhang K., Tu E., Gao Q., Chen W., Zhang C. and Zhang Y.

Regulatory T cells in cardiovascular diseases. Nat Rev Cardiol. 2016;13:167-79.

186. Libby P., Lichtman A. H. and Hansson G. K. Immune effector mechanisms implicated in

atherosclerosis: from mice to humans. Immunity. 2013;38:1092-104.

187. Li Miao, Lohmüller Theobald and Feldmann Jochen. Optical Injection of Gold

Nanoparticles into Living Cells. Nano Lett. 2015;15:770-775.

188. Cui Yi, Wang Xiaolei, Ren Wen, Liu Jing and Irudayaraj Joseph. Optical Clearing Delivers

Ultrasensitive Hyperspectral Dark-Field Imaging for Single-Cell Evaluation. ACS Nano.

2016;10:3132-3143.

189. Xie Xueping, Liao Jinfeng, Shao Xiaoru, Li Qianshun and Lin Yunfeng. The Effect of

shape on Cellular Uptake of Gold Nanoparticles in the forms of Stars, Rods, and Triangles. Sci

Rep. 2017;7:3827.

190. Connor Ellen E., Mwamuka Judith, Gole Anand, Murphy Catherine J. and Wyatt Michael

D. Gold Nanoparticles Are Taken Up by Human Cells but Do Not Cause Acute Cytotoxicity. Small.

2005;1:325-327.

191. Chithrani B Devika, Ghazani Arezou A and Chan Warren CW. Determining the size and

shape dependence of gold nanoparticle uptake into mammalian cells. Nano Lett. 2006;6:662-668.

192. Schmitt Joseph M and Kumar Gitesh. Optical scattering properties of soft tissue: a discrete

particle model. Appl Opt. 1998;37:2788-2797.

193. Hu Jie, Sanz-Rodríguez Francisco, Rivero Fernando, Rodríguez Emma Martín, Torres Río

Aguilar, Ortgies Dirk H, Solé José García, Alfonso Fernando and Jaque Daniel. Gold nanoshells:

Contrast agents for cell imaging by cardiovascular optical coherence tomography. Nano Res.

2018;11:676-685.

194. Hara Tetsuya and Jaffer Farouc A. Intravascular NIRF Molecular Imaging Approaches in

Coronary Artery Disease. Curr Cardiovasc Imaging Rep. 2016;9:13.

195. Lee Sunki, Lee Min Woo, Cho Han Saem, Song Joon Woo, Nam Hyeong Soo, Oh Dong

Joo, Park Kyeongsoon, Oh Wang-Yuhl, Yoo Hongki and Kim Jin Won. Fully integrated high-

speed intravascular optical coherence tomography/near-infrared fluorescence structural/molecular

imaging in vivo using a clinically available near-infrared fluorescence–emitting indocyanine green

to detect inflamed lipid-rich atheromata in coronary-sized vessels. Circulation: Cardiovascular

Interventions. 2014;7:560-569.

References

143

196. Kim Ji Bak, Park Kyeongsoon, Ryu Jiheun, Lee Jae Joong, Lee Min Woo, Cho Han Saem,

Nam Hyeong Soo, Park Ok Kyu, Song Joon Woo and Kim Tae Shik. Intravascular optical imaging

of high-risk plaques in vivo by targeting macrophage mannose receptors. Sci Rep. 2016;6:22608.

197. Diao Shuo, Blackburn Jeffrey L, Hong Guosong, Antaris Alexander L, Chang Junlei, Wu

Justin Z, Zhang Bo, Cheng Kai, Kuo Calvin J and Dai Hongjie. Fluorescence imaging in vivo at

wavelengths beyond 1500 nm. Angew Chem. 2015;127:14971-14975.

198. Robinson Joshua T, Hong Guosong, Liang Yongye, Zhang Bo, Yaghi Omar K and Dai

Hongjie. In vivo fluorescence imaging in the second near-infrared window with long circulating

carbon nanotubes capable of ultrahigh tumor uptake. J Am Chem Soc. 2012;134:10664-10669.

199. Antaris Alexander L, Chen Hao, Cheng Kai, Sun Yao, Hong Guosong, Qu Chunrong, Diao

Shuo, Deng Zixin, Hu Xianming and Zhang Bo. A small-molecule dye for NIR-II imaging. Nat

Mater. 2016;15:235-242.

200. Bu Xiaobing, Chen Huaping, Gai Hongwei, Yang Ronghua and Yeung Edward S.

Scattering Imaging of Single Quantum Dots with Dark-Field Microscopy. Anal Chem.

2009;81:7507-7509.

201. Schmitt Joseph M. Optical coherence tomography (OCT): a review. IEEE J Sel Top

Quantum Electron. 1999;5:1205-1215.

202. Leatherdale Catherine A, Woo W-K, Mikulec Frederic V and Bawendi Moungi G. On the

absorption cross section of CdSe nanocrystal quantum dots. The Journal of Physical Chemistry B.

2002;106:7619-7622.

203. Yu Pingrong, Beard Matthew C, Ellingson Randy J, Ferrere Suzanne, Curtis Calvin,

Drexler John, Luiszer Fred and Nozik Arthur J. Absorption cross-section and related optical

properties of colloidal InAs quantum dots. The Journal of Physical Chemistry B. 2005;109:7084-

7087.

204. Dong Ning-Ning, Pedroni Marco, Piccinelli Fabio, Conti Giamaica, Sbarbati Andrea,

Ramírez-Hernández Juan Enrique, Maestro Laura Martínez, Iglesias-de la Cruz Maria Carmen,

Sanz-Rodriguez Francisco and Juarranz Angeles. NIR-to-NIR two-photon excited CaF2: Tm3+,

Yb3+ nanoparticles: multifunctional nanoprobes for highly penetrating fluorescence bio-imaging.

ACS Nano. 2011;5:8665-8671.

205. Rocha Ueslen, Kumar Kagola Upendra, Jacinto Carlos, Villa Irene, Sanz‐Rodríguez

Francisco, Juarranz Angeles, Carrasco Elisa, van Veggel Frank CJM, Bovero Enrico and Solé José

García. Neodymium‐Doped LaF3 Nanoparticles for Fluorescence Bioimaging in the Second

Biological Window. Small. 2014;10:1141-1154.

206. Ruiz Diego, del Rosal Blanca, Acebrón María, Palencia Cristina, Sun Chen, Cabanillas‐

González Juan, López‐Haro Miguel, Hungría Ana B, Jaque Daniel and Juarez Beatriz H.

Ag/Ag2S Nanocrystals for High Sensitivity Near‐Infrared Luminescence Nanothermometry. Adv

Funct Mater. 2017;27:1604629.

207. Quillard Thibaut and Libby Peter. Molecular imaging of atherosclerosis for improving

diagnostic and therapeutic development. Circ Res. 2012;111:231-244.

Documents

Contrast agents for cardiovascular optical imaging at